Pscan Pro En

Agisoft PhotoScan User Manual

Professional Edition, Version 0.8.3

Agisoft PhotoScan User Manual: Professional Edition, Version 0.8.3

Publication date 2011Copyright © 2011 AgiSoft LLC

iii

Table of ContentsOverview ......................................................................................................................... iv

How it works ............................................................................................................ ivAbout the manual ...................................................................................................... iv

1. Installation ..................................................................................................................... 1System requirements ................................................................................................... 1OpenCL acceleration ................................................................................................... 1Installation procedure .................................................................................................. 1Restrictions of the trial mode ........................................................................................ 2

2. Capturing photos ............................................................................................................ 3Basic rules ................................................................................................................ 3Capturing scenarios ..................................................................................................... 3Restrictions ............................................................................................................... 4

3. General workflow ........................................................................................................... 6Loading photos .......................................................................................................... 6Aligning photos .......................................................................................................... 7Building model geometry ............................................................................................. 8Building model texture ................................................................................................ 9Saving intermediate results ......................................................................................... 10Exporting results ....................................................................................................... 11

4. Advanced use ............................................................................................................... 14Splitting project ........................................................................................................ 14Camera calibration .................................................................................................... 16Using masks ............................................................................................................ 18Editing model geometry ............................................................................................. 20Setting coordinate system ........................................................................................... 22Performing measurements ........................................................................................... 26

A. Graphical User Interface ................................................................................................ 28Application Window ................................................................................................. 28Menu Commands ...................................................................................................... 30Toolbar Buttons ........................................................................................................ 33

B. Troubleshooting ............................................................................................................ 35Photo alignment succeeds, but the resulting camera positions appear to be wrong .................. 35Reconstructed geometry appears to be cut and some important parts are missing ................... 35The photos included in the project file can't be opened and operations from the Workflowmenu fail ................................................................................................................. 36

iv

OverviewAgisoft PhotoScan is an advanced image-based 3D modeling solution aimed for creating the professionalquality 3D content from still images. Based on the latest multi-view 3D reconstruction technology, itoperates with arbitrary images and is efficient in both controlled and uncontrolled conditions. Photos canbe taken from any positions, providing that an object to be reconstructed is visible on at least two photos.Both image alignment and 3D model reconstruction are fully automated.

How it worksIn general the final goal of photographs processing by PhotoScan is to build the textured 3D model. Theprocedure of photographs processing and 3D model construction comprises three main stages.

1. First stage is the photographs alignment. At this stage PhotoScan searches for common points on pho-tographs and matches them, as well as it finds positions of the camera for each picture and refines thecamera calibration parameters. As a result the sparse point cloud and the set of camera positions areformed.

The point cloud represents the results of photos alignment and will not be directly used in the further3D model construction procedure (except the point cloud based reconstruction method). However itcan be exported for further usage by external tools. For instance the point cloud model can be used inthe 3D editor as a reference.

On the contrary the set of camera positions will be required for further 3D model construction byPhotoScan.

2. Next stage is building the geometry. Based on the estimated camera positions and pictures themselvesPhotoScan builds the 3D polygon mesh, representing the object surface. The four algorithmic methodsavailable in PhotoScan can be applied for 3D mesh generation: Arbitrary - Smooth, Arbitrary - Sharp,Height field - Smooth and Height field - Sharp methods. Additionally there is a Point Cloud basedmethod for fast geometry generation based on the sparse point cloud alone.

After building the mesh it may be necessary to edit it. Some corrections, such as mesh decimation, re-moval of detached components, closing of holes in the mesh and so on can be performed by PhotoScan.For more complex editing you have to engage external 3D editor tools. PhotoScan allows to export themesh, edit it by another software and import it back.

3. After the geometry (i.e. the mesh) is generated, it may be textured and/or used for orthophoto generation.Several texturing modes are available in PhotoScan, they are described in the corresponding sectionsof this manual.

About the manualBasically, the described sequence covers most of the model processing needs. All these operations arecarried out automatically according to the parameters set by user. The instructions how to get through theseoperations and descriptions of the parameters controlling each step are explained in the correspondingsections of the Chapter 3, General workflow.

However, in some cases additional actions may be required to get the desired results. For instance picturestaken using uncommon lenses such as fish-eyes may require preliminary calibration of optical system pa-rameters. In some capturing scenarios additional masking of certain regions of the photos may be requiredto exclude them from the calculations. After you built up the model you may set a reference system for

Overview

v

it and make additional calculations, such as estimation of the surface area and volume of the model. Allthese advanced functions are described in the Chapter 4, Advanced use.

Building of a 3D model may take quite a long time. At any stage of the process PhotoScan allows you toexport the results obtained and to save the intermediate data in a form of project files. If you are not familiarwith the concept of projects, its brief description is given at the end of the Chapter 3, General workflow.

Also in the manual you will find instructions on the PhotoScan installation procedure and basic rules fortaking "good" photographs, i.e. pictures that allow to get most necessary information for 3D reconstruction.

1

Chapter 1. Installation

System requirementsMinimal configuration

• Windows (XP or later) or Mac OS X (Snow Leopard or later), 32 or 64 bit

• Intel Core 2 Duo processor or equivalent

• 2GB of RAM

Recommended configuration

• Windows (XP or later) or Mac OS X (Snow Leopard or later), 64 bit

• Intel Core i7 processor

• 12GB of RAM

The number of photos that can be processed by PhotoScan depends on the available RAM and reconstruc-tion parameters used. Assuming that a single photo resolution is of the order of 10 MPx, 2GB RAM issufficient to make a model based on 20 to 30 photos. 12GB RAM will allow to process up to 200-300photographs.

OpenCL accelerationPhotoScan supports an accelerated geometry reconstruction using the graphics hardware (GPU).

NVidiaGeForce 8xxx series and later.

ATIRadeon HD 5xxx series and later.

Note

• OpenCL acceleration can be enabled using OpenCL tab in the Preferences dialog box.

Installation procedure

Installing PhotoScan on Microsoft WindowsTo install PhotoScan simply run the downloaded msi file and follow the instructions.

Installing PhotoScan on Mac OS XOpen the downloaded dmg image and drag the PhotoScan application to the desired location on your harddrive.

Installation

2

Installing PhotoScan on Debian/UbuntuUnpack the downloaded archive with a program distribution to the desired location on your hard drive.Start PhotoScan by running photoscan.sh script from the program folder.

Restrictions of the trial modeWhen PhotoScan is downloaded and installed on your computer you can run it either in the trial modeor in the full-function mode. At every run until you enter the serial number it will show the registrationbox, offering the two options: (1) use PhotoScan in the trial mode or (2) enter the serial number to confirmthe purchase. The first choice is set by default, so if you are still exploring PhotoScan click the Continuebutton and PhotoScan will start in the trial mode.

The usage of PhotScan in the trial mode is not limited in time. However, several functions are not availablein the trial mode. These functions include:

• saving the project;

• exporting any reconstruction results (you may only view 3D model on the screen)

To use PhotoScan in the full-function mode you have to purchase it. After the purchase you will get theserial number. Enter this number in the registration box next time you start PhotoScan. From now on theregistration box will not appear again and you will have full access to all the PhotoScan functions.

3

Chapter 2. Capturing photosBefore you load your photographs to PhotoScan you need to take them and select ones suitable for 3Dmodel construction.

Photographs can be taken by any standard digital camera, as long as you follow some specific capturingguidelines. This section explains the general principles of taking and selecting pictures that provide themost appropriate data for 3D model generation.

Basic rules• Use a digital camera with a reasonably high resolution (5 MPix or more).

• Wide angle lenses suits better for reconstructing spatial relations between objects than telephoto ones.

• Avoid un-textured and flat objects or scenes.

• Avoid shiny and transparent objects.

• Avoid unwanted foregrounds and moving objects as much as possible.

• Shoot shiny objects under a cloudy sky.

• Shoot pictures of the scene with a lot of overlap.

• Capture most important scene content from multiple viewpoints (3 or more).

• Do not crop or geometrically transform the images.

• More photos is better than not enough.

• Spending some time planning your shot might be very useful.

• If you intend to set a reference coordinate system for reconstructed 3D model, make sure to identifyor place on ground marks that will be used to set reference coordinate system and scale. Measure areference distance.

• Make sure to study the following schemes and read the list of restrictions before you get out for shootingphotographs.

Capturing scenariosThe following figures represent several typical capturing scenarios:

Capturing photos

4

Facade (Incorrect) Facade (Correct)

Interior (Incorrect) Interior (Correct)

Isolated Object (Incorrect) Isolated Object (Correct)

RestrictionsIn some cases it might be very difficult or even impossible to build a correct 3D model from a set ofpictures. The short list of typical reasons for such photographs unsuitability is given below.

Modifications of photographsPhotoScan can process only unmodified photos as they were taken by a digital photo camera. Processingthe photos which were manually cropped or geometrically warped is likely to fail or produce highly inac-curate results. Photometric modifications do not affect the reconstruction results.

Capturing photos

5

Lack of EXIF dataWhen estimating the field of view for each photo PhotoScan uses the information saved in the EXIF partof each picture. If EXIF data are available you may expect to get the best possible 3D reconstruction.However 3D scene may also be reconstructed in the absence of EXIF data. In this case PhotoScan assumesthat the 35mm focal length equivalent equals to 50 mm and it tries to align the photos in accordance withthis assumption. If the correct focal length value differs significantly from 50 mm, the alignment can giveincorrect results or even fail. In such cases it is required to specify the initial camera calibration manually.

The details of necessary EXIF tags and instructions for manual setting of the calibration parameters aregiven in the Camera calibration section

Lens distortionThe distortion of the lens being used to capture photos should be well modeled using the Brown's distortionmodel. Otherwise it is almost impossible to build a precise 3D model. Fish eyes and ultra-wide angle lensesare usually badly modeled by the distortion model used and may produce inaccurate reconstructions.

6

Chapter 3. General workflowThe processing of images by PhotoScan includes the following main steps:

• loading photos in PhotoScan;

• inspecting loaded images, removing unnecessary images;

• aligning photos;

• building 3D model;

• editing 3D model;

• exporting results.

If you are using PhotoScan in the full-function (not the trial) mode the intermediate results of the imageprocessing may be saved at any stage in the form of project files and may be used later. The concept ofprojects and project files is briefly explained in the Saving intermediate results section.

The list above represents all the necessary steps involved in the construction of a textured 3D modelfrom your photos. Some additional tools which you may find to be useful are described in the Chapter 4,Advanced use.

Loading photosBefore starting any operation it is necessary to point out what photos will be used as a source for 3Dreconstruction. In fact photographs themselves are not loaded into PhotoScan unless they are needed. So,when you "load photos" you only indicate photographs that will be used for further processing.

To load a set of photos

1.Select Add Photos... command from the Workflow menu or click Add Photos toolbar button.

2. In the Add Photos dialog box, browse to the folder, containing the images, and select files to beprocessed. Then click Open button.

3. Selected photos will appear in the Workspace pane.

Note

• PhotoScan accepts the following image formats: JPEG, TIFF, PNG, BMP, PPM, CR2 and JPEGMulti-Picture Format (MPO). Photos of any other format will not be shown in the Add Photosdialog box. To work with such photos you will need to convert them in one of the supportedformats.

If you have loaded some unwanted photos, you can easily remove them at any time.

To remove unwanted photos

1. In the Workspace pane select the photos to be removed.

2. Right-click on the selected photos and select Remove Items command from the opened context menu,

or click Remove Items toolbar button. The selected photos will be removed from the working set.

General workflow

7

Inspecting the loaded photosThe loaded photos are displayed in the Workspace pane along with flags reflecting their status.

The following flags may appear next to the photo name:

NC (Not calibrated)Notifies that the EXIF data available is not sufficient to estimate the camera focal length. In this casePhotoScan assumes that the corresponding photo was taken using 50mm lens (35mm film equivalent).If the actual focal length differs significantly from this value, manual calibration may be required.More details on manual camera calibration are available in the Camera calibration section.

NA (Not aligned)Notifies that external camera orientation parameters were not estimated for the current photo yet.

Images loaded to PhotoScan will not be aligned until you perform the next step - photos alignment.

Aligning photosWhen photos are loaded to PhotoScan they need to be aligned. At this stage PhotoScan finds the cameraposition for each photo and builds the point cloud model.

To align a set of photos

1. Select Align Photos... command from the Workflow menu.

2. In the Align Photos dialog box, select the desired alignment options. Click OK button when done.

3. The progress dialog box will appear displaying the current processing status. To cancel processingclick the Cancel button.

After the alignment is complete, the computed camera positions and a sparse point cloud will be displayed.You can inspect the alignment results and remove the incorrectly positioned photos, if any.

The point cloud and estimated camera positions can be exported for processing by another software ifneeded.

Alignment parametersThe following parameters control the photo alignment procedure and can be modified in the Align Photosdialog box:

AccuracyHigher accuracy setting helps to obtain more accurate camera position estimates. Lower accuracysetting can be used to get the rough camera positions in the shorter time.

Pair preselectionThe alignment process of large photo sets may take a long time. A signficant portion of this time isspent for matching of detected features across the photos. Image pair preselection option can speedup this process by selection of a subset of image pairs to be matched. In the Generic preselectionmode the overlapping pairs of photos are selected by matching photos using lower accuracy settingfirst. In the Ground Control preselection mode the overlapping pairs of photos are selected basedon the measured camera locations (if present).

General workflow

8

Constrain features by maskWhen this option is enabled, features detected in the masked image regions are discarded. For addi-tional information on the usage of masks please refer to the Using masks section.

Building model geometry3D model reconstruction is a computationally intensive operation and can take a long time, depending onthe quantity and resolution of loaded photos. It is recommended to build the model with the lowest qualityfirst to estimate the applicability of the chosen reconstruction method, and then to recompute the resultsusing the higher quality setting. It is also recommended to save your project before starting to build thegeometry.

To build a 3D model

1.Check the reconstruction volume bounding box. If required, adjust the bounding box using the Re-

size Region and Rotate Region toolbar buttons. Rotate the bounding box and then drag cornersof the box to the desired positions. When using Height field reconstruction method, the redside of the bounding box will define the reconstruction plane. Make sure that the bounding box iscorrectly oriented in this case.

2. Select the Build Geometry... command from the Workflow menu.

3. In the Build Geometry dialog box, select the desired reconstruction parameters. Click OK buttonwhen done.


Reconstruction methodsAgisoft PhotoScan supports several reconstruction methods and settings, which help to produce optimalreconstructions for a given data set.

Object type

ArbitraryArbitrary object type can be used for modeling any kind of objects. It should be selected for closedobjects, such as statues, buildings, etc. It doesn't make any assumptions on the type of the objectmodeled, which comes at a cost of higher memory usage.

Height fieldThe Height field object type is optimized for modeling of planar surfaces, such as terrains orbas-reliefs. It should be selected for aerial phototography processing as it requires lower amount ofmemory and can process larger data sets.

Geometry type

SharpThe reconstructions produced with Sharp geometry type selection are more accurate and do not in-troduce extra geometry, like hole filling. Manual hole filling is usually required at the post processingstep.

General workflow

9

SmoothSmooth geometry type selection produces watertight reconstructions with no or little holes on re-sulting surface. Large areas of extra geometry could be generated with this method, but they couldbe easily removed later using selection and cropping tools. Smooth setting is recommended for or-thophoto generation.

Reconstruction parametersQuality

Specifies the desired reconstruction quality. Higher quality settings can be used to obtain more detailedand accurate geometry, but require longer time for processing. Point Cloud quality setting is usedfor fast 3D model generation based on the sparse point cloud alone and is available for Smoothgeometry type only.

Face countSpecifies the maximum face count in the final mesh. 0 - if no decimation is required.

Filter thresholdSpecifies the maximum face count of small connected components to be removed after surface recon-struction (in percents of the total face count). The 0 value disables connected component filtering.

Hole threshold (Height field methods only)The maximum size of holes to be filled after surface reconstruction (in percents of the total surfacearea). The 0 value disables automatic hole filling.

Note

• PhotoScan tends to produce 3D models with excessive geometry resolution, so it is recommendedto perform mesh decimation after geometry computation. More information on mesh decimationand other 3D model geometry editing tools is given in the Editing model geometry section.

Building model textureTo generate 3D model texture

1. Select Build Texture... command from the Workflow menu.

2. Select the desired texture generation parameters in the Build Texture dialog box. Click OK buttonwhen done.


Texture mapping modesThe texture mapping mode determines how the object texture will be packed in the texture atlas. Selectionof a proper texture mapping mode will help to obtain the optimal texture packing and thus better visualquality of the final model.

GenericThe default mode is the Generic mapping mode; it is able to parameterize the texture atlas for anarbitrary geometry. It doesn't make any assumptions regarding the type of the scene to be processed,and tries to create as uniform texture as possible.

General workflow

10

Adaptive orthophotoIn the Adaptive orthophoto mapping mode the object surface is split into the flat part andvertical regions. The flat part of the surface is textured using the orthographic projection, while verticalregions are textured separately to maintain the accurate texture representation in such regions. TheAdaptive orthophoto mapping mode tends to produce more compact texture representationfor nearly planar scenes while maintaining good texture quality for vertical surfaces, such as wallsof the buildings.

OrthophotoIn the Orthophoto mapping mode the whole object surface is textured in the orthographic pro-jection. The Orthophoto mapping mode produces even more compact texture representation thenAdaptive orthophoto mode at the expense of texture quality in vertical regions.

Single photoThe Single photo mapping mode allows to generate texture from a single photo. The photo tobe used for texturing can be selected from Texture from list.

Keep uvThe Keep uv mapping mode generates texture atlas using current texture parameterization. It canbe used to rebuild texture atlas using different resolution or to generate the atlas for the modelparametrized in the external software.

Texture generation parametersThe following parameters control various aspects of texture atlas generation:

Texture from (Single photo mapping mode only)Specifies the photo to be used for texturing. Applicable only in the Single photo mapping mode.

Blending mode (not used in Single photo mode)Selects the way how pixel values from different photos are combined in a final texture.

Average - uses the average value of all pixels from individual photos.

Mosaic - gives more quality for orthophoto and texture atlas than Average mode as long as it doesnot mix image details between overlapping photos but uses more appropriate photo. Mosaic textureblending mode is especially useful for orthophoto generation based on approximate geometric model.

Max Intensity - the photo which has maximum intensity of the corresponding pixel is selected.

Min Intensity - the photo which has minimum intensity of the corresponding pixel is selected.

Atlas widthSpecifies the width of the texture atlas in pixels.

Atlas heightSpecifies the height of the texture atlas in pixels.

Saving intermediate resultsCertain stages of 3D model reconstruction may take a long time. The whole chain of operations couldeasily occupy 4-6 hours when building a model from hundreds of photos. It is not always possible to finishall the operations in one run. PhotoScan allows to save intermediate results in a project file.

PhotoScan project files may contain the following information:

General workflow

11

• List of loaded photographs with reference paths to the image files.

• Photo alignment data such as information on camera positions, point cloud model and set of refinedcamera calibration parameters for each photo.

• Reconstructed 3D model with any changes made by user. This includes geometry and texture if theywere built.

• Structure of the project, i.e. number of chunks in the project and their content.

You may save the current state of your work at the end of any processing stage and return to it later, simplyby loading the corresponding file. Project files may also be applied as backup files or used to save differentversions of your models.

Project files use relative paths to reference original photos. So, when moving or copying the project fileto another location you also need to move or copy photographs with all the folder structure involved. Inother case PhotoScan will load the file including the reconstructed model correctly, but all operations,requiring source images will fail.

Exporting resultsPhotoScan supports export of processing results in various representations. Sparse point clouds, 3D mod-els, orthophotos and digital elevation models may be generated depending on the user requirements.

Sparse point cloud may be exported right after photo alignment is completed. All other export options areavailable after the geometry is built.

In some cases editing the model geometry in the external software may be required. PhotoScan supportsexporting the model for editing in such software and then importing it back, as it is described in the Editingmodel geometry section of the manual.

All the export commands are available from the File menu.

Point cloud exportExports the point cloud model and/or estimated camera positions in one of the following formats:

• Wavefront OBJ

• Stanford PLY

• CHAN file format

• Boujou TXT file format

• Bundler OUT file format

• Omega Phi Kappa text file format

• PhotoScan structure file format (XML based)

Note

• Saving color information of the point cloud is supported by the Bundler and PLY file formatsonly.

General workflow

12

3D model exportExports the 3D model in one of following formats:

• Wavefront OBJ

• 3DS file format

• VRML

• Stanford PLY

• COLLADA

• Autodesk DXF

• U3D

There is also an option of direct embedding of U3D models in a single page PDF-file. The format choicedepends on the further usage of the model.

Some file formats save texture image in a separate file. This file should be kept in the same directory as themain file describing the geometry. If the texture atlas was not built only the model geometry is exported.

Orthophoto exportOrthophoto export is normally used for generation of high resolution imagery based on the source photosand reconstructed geometry. The most common application is aerial photographic survey processing, butit may be also useful when a detailed view of the object is required. Orthophoto is often accompanied bythe digital elevation model (see the next section).

In order to export orthophoto in the correct orientation it is necessary to set the coordinate system for themodel first. As far as PhotoScan uses original images to build an orthophoto there is no need to build thetexture atlas.

The following formats are supported for orthophoto export:

• JPEG

• PNG

• TIFF

• GeoTIFF

• Multiresolution Google Earth KML mosaic.

DEM exportDigital elevation models (DEMs) represent the model surface as a regular grid of height values and are of-ten used for aerial photographic survey data. Digital elevation model may be combined with the orthophototo produce 3D model of an area.

Properly defined coordinate system is required for DEM export.

The following formats are supported for DEM export:

General workflow

13

• GeoTIFF elevation data

• Arc/Info ASCII Grid (ASC)

• Band interlieved file format (BIL)

• XYZ file format

14

Chapter 4. Advanced useSplitting project

In some cases it is very hard or even impossible to generate 3D model of the whole object in one go. Thiscould happen for instance if the total amount of photographs is too large to be processed. To overcomethis PhotoScan offers the possibility of splitting the set of photos in several separate "chunks" within theproject. The alignment of photos, building geometry and forming the texture atlas may be performed oneach chunk separately and then the resulting 3D models may be combined together.

Working with chunks is no more difficult than using PhotoScan following the general workflow. In fact,in PhotoScan always exists at least one active chunk and all the 3D model processing workflow operationsare applied to this chunk.

All you need to know to work with several chunks is how to create chunks and how to combine resulting3D models from separate chunks into the one.

Creating a chunk

To create new chunk click on the Add Chunk toolbar button on the Workspace pane or select AddChunk command from the Workspace context menu (available by right-clicking on the root element inthe Workspace pane).

After the chunk is created you may load photos in it, align them, generate mesh surface model, build textureatlas, export the models at any stage and so on. The models in the chunks are not linked with each other.

To move photos from one chunk to another simply select them in the list of photos on the Workspace pane,and then drag and drop in the desirable chunk.

Working with chunksAll operations within the chunk are carried out following the common workflow: loading photographs,aligning them, building geometry model, building texture atlas, exporting 3D model and so on.

Note that all these operations are applied to the active chunk. When a new chunk is created it is activatedautomatically. Save project operation saves the content of all chunks.

To set another chunk as active

1. Right-click on the chunk title in the Workspace pane.

2. Select Set Active command from the context menu.

To remove chunk

1. Right-click on the chunk title in the Workspace pane.

2. Select Remove Items command from the context menu.

Aligning models from several chunksAfter the "partial" 3D models are built in several chunks they can be merged together. Before mergingthe models they need to be aligned.

Advanced use

15

To align models from separate chunks

1. Select Align Chunks command from the Workflow menu.

2. In the Align Chunks dialog box select chunks to be aligned, set reference chunk and the desiredalignment options. Click OK button when done.


Note

• Chunks alignment can be performed if at least two chunks contain aligned photos.

Chunks alignment parameters

The following parameters control the chunks alignment procedure and can be modified in the Align Chunksdialog box:

MethodDefines the chunks alignment method. Point based method alignes chunks by matching photosacross the different chunks. Marker based method uses markers as common points between dif-ferent chunks. The details on using markers are available in the Setting coordinate system section.

Accuracy (Point based alignment only)Higher accuracy setting helps to obtain more accurate chunk alignment results. Lower accuracy settingcan be used to get the rough chunk alignment in the shorter time.

Preselect image pairs (Point based alignment only)The alignment process of many chunks may take a long time. A signficant portion of this time is spentfor matching of detected features across the photos. Image pair preselection option can speed up thisprocess by selection of a subset of image pairs to be matched.

Constrain features by mask (Point based alignment only)When this option is enabled, features detected in the masked image regions are discarded. For addi-tional information on the usage of masks refer to the Using masks section.

Merging models from several chunksAfter alignment is complete the separate models can be merged into a single model.

To merge models

1. Select Merge Chunks command from the Workflow menu.

2. In the Merge Chunks dialog box select chunks to be merged and the desired merging options. ClickOK button when done.

3. PhotoScan will merge the separate chunks into one. The merged chunk will be displayed in the projectcontent list on Workspace pane.

Chunks merging parameters

The following parameters control the chunks merging procedure and can be modified in the Merge Chunksdialog box:

Advanced use

16

Merge modelsDefines if models from the selected chunks are merged.

Merge markersDefines if markers from the selected chunks are merged (only markers with the same labels wouldbe merged).

Chunks merging result (i.e. photos, points and geometry) will be stored in the new chunk and it may betreated as common chunk (e.g. textured and/or exported).

Batch processingPhotoScan allows to perform general workflow operations with multiple chunks automatically. It is usefulwhen dealing with a large number of chunks to be processed.

Batch processing can be applied to all chunks in the Workspace, to unprocessed chunks only, or to thechunks selected by the user.

Batch processing can perform the following operations:

• Align photos

• Build geometry

• Build texture

• Save project after every completed operation

To start batch processing

1. Select Batch Process... command from the Workflow menu.

2. Click Add to add the desired processing stages.

3. In the Add Job dialog select the kind of operation to be performed, the list of chunks it should beapplied to, and desired processing parameters. Click OK button when done.

4. Repeat the previos steps to add other processing steps as required.

5. Click OK button to start processing.

Camera calibrationWhile carrying out photos alignment PhotoScan estimates both internal and external camera orientationparameters, including nonlinear radial distortions. For a successful estimation the information on approx-imate focal length values is required. Normally this data is extracted automatically from the EXIF meta-data, or the 50mm focal length (35mm film equivalent) is assumed if the EXIF information is insufficient.

In some cases, the 50mm guess can differ too much from the actual focal length. This can lead to the failureof the alignment process. In such cases it is required to specify the initial camera calibration manually.

To specify camera calibration manually

1. Select Camera Calibration... command from the Tools menu.

Advanced use

17

2. In the Camera Calibration dialog box, select Initial viewing mode. Select the photos to be cali-brated.

3. Modify the calibration parameters displayed in the corresponding edit boxes. Normally only fx, fy,cx and cy values should be corrected.

4. Press Apply button to set the calibration. You will notice that the selected photos will be marked withthe C flag (Custom calibration).

5. Press Close button to close the camera calibration dialog.

Calibration parametersfx, fy

Focal length in x- and y-dimensions measured in pixels.

cx, cyPrincipal point coordinates, i.e. coordinates of lens optical axis interception with sensor plane.

skewSkew transformation coefficient.

k1, k2, k3Radial distortion coefficients.

p1, p2Tangential distortion coefficients.

How to set up appropriate valuesIn the case of missing EXIF data or too large lens distortions manual setting of lens calibration may berequired. This may happen if you used an old manual lens which cannot be detected by the camera or ifthe photos were taken by the film camera and digitized later. Some hints on selecting suitable calibrationparameters are listed below:

• It is very unlikely that skew transformation coefficient will have non zero value. So its almost alwayssafe to set skew parameter equal to zero.

• Unless you have used lenses with high distortion, such as fish eyes, the same applies to distortion coef-ficients. Leave it zero and PhotoScan will estimate them on its own. If ultra wide or fisheye lens wereused to capture the photos you may try to estimate them using Agisoft Lens utility for lens calibration(http://www.agisoft.ru/products/lens/).

• fx and fy coefficients are normally equal. They may be different if the pixels of recording media werenot square, which is a very rare case. If you know the focal length of the lens, physical dimension of theoptical sensor and size of picture in pixels you may easily calculate these coefficients as: fx = focal lengthin mm * x-dimension of sensor in pixels / x-dimension of sensor in mm. Similar formula applies to fy.

• For conventional (non tilt and shift) lenses the optical axis of the lens is likely to intercept the opticalsensor plane in its geometric center. So for cx and cy coefficient the coordinates of the image centerwould be the good guess.

Automatic refinement of camera calibration parametersBy default PhotoScan considers specified camera calibration parameters as the initial guess, and refinesthem later during the photo alignment. That is generally a desirable behaviour. However in the cases when

http://www.agisoft.ru/products/lens/

Advanced use

18

the camera calibration is known precisely (like in case of a metric camera), it may be required to protectcamera calibration parameters from optimization. To fix the camera calibration parameters select the Fixcalibration check box in the Camera Calibration dialog.

Using masks

Overview

Masks are used in PhotoScan to specify the areas on the photos which can otherwise be confusing tothe program or lead to incorrect reconstruction results. Masks can be applied at the following stages ofprocessing:

• Alignment of the photos

• Building 3D model geometry

• Building 3D model texture

Alignment of the photos

Masked areas can be excluded during feature point detection. Thus, the objects on the masked parts of thephotos are not taken into account while estimating camera positions. This is important in the setups, wherethe object of interest is not static with respect to the scene, like when using a turn table to capture the photos.

Masking may be also useful when the object of interest occupies only a small part of the photo. In thiscase a small number of useful matches can be filtered out mistakenly as a noise among a much greaternumber of matches between background objects.

Building 3D model

While building the model, masked areas are not used in the surface generation process. Masking can beused to reduce the resulting model complexity, by eliminating the areas on the photos that are not of interest.

Masked areas are always excluded from processing during surface reconstruction and texture generationstages.

Let's take for instance a set of photos of some object. Along with an object itself on each photo somebackground areas are present. These areas may be useful for more precise camera positioning, so it isbetter to use them while aligning the photos. However, impact of these areas at the building geometrystage is exactly opposite. If they are used for building geometry the resulting model will contain objectof interest and its background. Background geometry will "consume" some part of mesh polygons thatwould be otherwise used for modeling the main object.

Advanced use

19

Setting the masks for such background areas allows to avoid this problem and increases the precision andquality of geometry reconstruction.

Building texture atlas

During texture atlas generation, masked areas on the photos are not used for texturing. Masking areas onthe photos that are occluded by outliers or obstacles helps to prevent the "ghosting" effect on the resultingtexture atlas.

Mask editingMasks in PhotoScan are represented by contours, bounding the certain image regions. Modification of thecurrent mask is performed by adding or subtracting selections. A selection is created by using one of thesupported selection tools and is not incorporated in the current mask until it is merged with a mask usingAdd Selection or Subtract Selection operations.

To edit the mask

1. Open the photo to be masked by double clicking its name in the Workspace pane. The photo will beopened in the main window. The existing mask will be displayed as a shaded region on the photo.

2. Select the desired selection tool and generate a selection.

3.Click on Add Selection toolbar button to add current selection to the mask, or Subtract Se-

lection to subtract the selection from the mask. Invert Selection button allows to invert currentselection prior to adding or subtracting it from the mask.

If photo has alpha channel it could be applied as a mask to one or more photos in Workspace.

To load mask from alpha channel

1. Open the photo containing alpha channel by double clicking on its name in the Workspace pane. Thephoto will be opened in the main window.

2. Select Import mask... command from Photo menu.

3. In the Import Mask dialog box select whether the mask should be imported to the current photo only,all photos in the chunk or to the entire Workspace. Click OK button when done.

Supported selection toolsThe following tools can be used for creating selections:

Rectangle selection tool

Rectangle selection tool is used to select large areas, or to cleanup the mask after other selection toolswere applied.

Intelligent scissors tool

Intelligent scissors is used to generate a selection by specifying its boundary. The boundary is formedby selecting a sequence of vertices with a mouse, which are automatically connected with segments. The

Advanced use

20

segments can be formed either by straight lines, or by curved contours snapped to the object boundaries.To enable snapping, hold Ctrl key while selecting the next vertex. To complete the selection, the boundaryshould be closed by clicking on the first boundary vertex.

Intelligent paint tool

Intelligent paint tool is used to "paint" a selection by the mouse, continuously adding small image regions,bounded by object boundaries.

Magic Wand tool

Magic Wand tool is used to select uniform areas of the image. To make a selection with a Magic Wandtool, click inside the region to be selected.

The range of pixel colors selected by Magic Wand is controlled by the tolerance value. At lower tolerancevalues the tool selects fewer colors similar to the pixel you click with the Magic Wand tool. Higher valuebroadens the range of colors selected.

Note

• The masks are generated individually for each image. If some object should be masked out, itshould be masked out on all photos, where that object appears.

• If your setup requires exactly the same masking for all input photos, you can use Propagate Maskcommand from the Photo menu to apply the current mask for all photos.

Editing model geometryThe following mesh editing tools are available in PhotoScan:

• Decimation tool

• Connected component filtering

• Manual face removal

More complex editing can be done in the external 3D editing tools. PhotoScan allows to export mesh andthen import it back for this purpose.

Note

• For face removal operations such as manual face removal and connected component filtering itis possible to undo the last mesh editing operation. There are Undo/Redo command in the Editmenu.

Please note that undo/redo commands are not supported for mesh decimation and this operationcan not be undone.

Decimation toolDecimation is a tool used to decrease the geometric resolution of the model by replacing high resolutionmesh with a lower resolution one, which is still capable of representing the object geometry with high

Advanced use

21

accuracy. PhotoScan tends to produce 3D models with excessive geometry resolution, so mesh decimationis usually a desirable step after geometry computation.

The high detailed models may contain hundreds of thousands polygons. While it is acceptable to workwith such a complex models in 3D editor tools, in most conventional tools like Adobe Reader or GoogleEarth high complexity of 3D models may noticeably decrease application performance. High complexityalso results in longer time required to build texture and to export model in pdf file format.

In some cases it is desirable to keep as much geometry details as possible like it is needed for scientific andarchive purposes. However, if there are no special requirements it is recommended to decimate the modeldown to 100 000 - 200 000 polygons for exporting in PDF, and to 100 000 or even less for displaying inGoogle Earth and alike tools.

To decimate 3D model

1. Select Decimate Mesh... command from the Tools menu.

2. In the Decimate Mesh dialog box specify the target number of faces, which should remain in the finalmodel. Click on the OK button to start decimation.

3. The progress dialog box will appear displaying the current processing status. To cancel processingclick on the Cancel button.

4. Texture atlas is discarded during decimation process. You will be prompted to rebuild texture atlasafter decimation is complete.

Connected component filteringIn some cases reconstructed geometry may contain the cloud of small isolated mesh fragments surroundingthe "main" model. Connected component filtering helps to remove small isolated mesh fragments, whichare usually reconstructed incorrectly or are unneeded.

To remove small isolated mesh fragments

1. Select Select Components... command from the Edit menu.

2. In the Select Components dialog box select the amount of isolated components to be selected usingthe slider. You can observe how the selection changes while dragging the slider. Click OK buttonto finalize the selection.

3.To remove the selected components select Delete Faces command from the Edit menu or click Delete Faces toolbar button (or simply press Del button on the keyboard).

Note that PhotoScan always selects the fragments starting from the smallest ones. If the model containsonly one component the selection will be empty.

Manual face removalUnnecessary and excessive sections of model geometry can be also removed manually.

To manually remove part of the faces

1.Select rectangle or circle selection tool using Rectangle Selection or Circle Selection toolbarbuttons.

Advanced use

22

2. Make the selection using the mouse. To add new faces to the current selection hold the Ctrl keyduring selection of additional faces.

3.To delete selected faces click the Delete Faces toolbar button. To crop selection to the selected

faces click the Crop Faces toolbar button or select Crop Faces command from the Edit menu.

To grow or shrink current selection

1. To grow current selection press PageUp key in the selection mode. To grow selection by even a largeramount, press PageUp while holding Shift key pressed.

2. To shrink current selection press PageDown key in the selection mode. To shrink selection by evena larger amount, press PageDown while holding Shift key pressed.

Editing mesh in the external program

To export mesh for editing in the external program

1. Select Export Model... command from the File menu.

2. In the Save As dialog box, specify the desired mesh format in the Save as type combo box. Select thefile name to be used for the model and click Save button.

3. In the opened dialog box specify additional parameters specific to the selected file format. Click OKbutton when done.

To import edited mesh

1. Select Import Mesh... command from the Tools menu.

2. In the Open dialog box, browse to the file with the edited model and click Open.

Note

• PhotoScan suppots loading models in Wavefront OBJ and Stanford PLY file formats only. Pleasemake sure to select one of these file formats when exporting model from the external 3D editor.

Setting coordinate systemMany applications require data with a defined coordinate system. Setting the coordinate system also pro-vides a correct scaling of the model allowing surface area and volume measurements and makes modelloading in geoviewers and geoinformatics software much easier. Some functionality like digital elevationmodel export is available only after the coordinate system is defined.

PhotoScan supports setting a coordinate system based on either ground control point (marker) coordinatesor camera coordinates. In both cases the coordinates are specified in the Ground Control dialog and canbe either loaded from the external file or typed in manually.

Setting coordinate system based on recorded camera positions is often used in aerial photography process-ing. However it may be also useful for processing photos captured with GPS enabled cameras. Placingmarkers is not required if recorded camera coordinates are used to initialize the coordinate system.

Advanced use

23

In the case when ground control points are used to set up the coordinate system the markers should beplaced in the corresponding locations of the scene.

Using camera positioning data for georeferencing the model is faster as manual marker placement is notrequired. On the other hand, ground control point coordinates are usually more accurate than telemetrydata, allowing for a more precise georeferencing.

Placing markersPhotoScan uses markers to specify locations within the scene. Markers are used for setting up a coordinatesystem, measuring distances within the scene as well as for marker based chunk alignment. Marker posi-tions are defined by their projections on the source photos. The more photos are used to specify markerposition the higher is accuracy of marker placement. To define marker location within a scene it shouldbe placed on at least 2 photos.

Note

• Marker placement is not required for setting the coordinate system based on recorded cameracoordinates. This section can be safely skipped if the coordinate system will be defined basedon recorded camera locations.

PhotoScan supports two approaches to marker placement: manual marker placement and guided markerplacement. In the manual approach the marker projections should be specified manually for each photowhere the marker is visible. Manual marker placement doesn't require 3D model and can be performedeven before photo alignment.

In the guided approach the marker projection is specified for the single photo only. PhotoScan automat-ically projects the corresponding ray onto the model surface and calculates marker projections on theremaining photos. Marker projections defined automatically on individual photos can be further refinedmanually. Reconstructed 3D model surface is required for the guided approach.

Guided marker placement usually speeds up the procedure of marker placement significantly and alsoreduces the chance of incorrect marker placement. It is recommended in most cases unless there are anyspecific reasons preveting its operation.

To place a marker using guided approach

1. Open the photo where the marker is visible by double clicking on its name.

2.Switch to the marker editing mode using Edit Markers toolbar button.

3. Right click on the photo at the point corresponding to the marker location.

4. Select Create Marker command from the context menu. New marker will be created and its projectionson the other photos will be automatically defined.

Note

• If the 3D model is not available or the ray at the selected point does not intersect with the modelsurface, the marker projection will be defined on the current photo only.

• Guided marker placement can be performed in the same way from the 3D view by clicking on

the corresponding point on the model surface using Create Marker tool. While the accuracy

Advanced use

24

of marker placement in the 3D view is usually much lower, it may be still useful for quicklylocating the photos observing the specified location on the model.

To place a marker using manual approach

1. Create marker instance using Add Marker command from the Chunk context menu (available by rightclicking on the chunk title in the Workspace pane).

2. Open the photo where the marker projection needs to be added by double clicking on its name.


4. Right click on the photo where the marker projection needs to be placed. From the context menu openPlace Marker submenu and select the marker instance previously created. The marker projection willbe added to the current photo.

5. Repeat the previous step to place marker projections on other photos as needed.

Defined marker locations can be refined later by dragging their projections on the corresponding photos.

To refine marker location

1. Open the photo where the marker is visible by double clicking on its name.


3. Move the marker projection to the desired location by dragging it using the left mouse button.

Note

• To list the photos where the marker locations are defined, select the corresponding marker on

the Workspace pane. The photos where the marker is placed will be marker with a icon onthe Photos pane.

PhotoScan automatically assigns default labels for each newly created marker. These labels can be changedusing the Rename... command from the marker context menu.

Assigning reference coordinatesTo reference the model the real world coordinates of at least 3 points in the scene should be specified. De-pending on the requirements, the model can be referenced using marker coordinates, camera coordinates,or both. Real world coordinates used for referencing the model along with the type of coordinate systemused are specified using the Gound Control dialog.

The model can be referenced using either local Euclidean coordinates, or geographic coordinates. Forgeoreferencing the model a wide range of various geographic projections is supported, including widelyused WGS84 coordinate system. Besides that, allmost all coordinate systems from the EPSG registry arewell supported as well.

The reference coordinates can be specified in one of the following ways:

• Loaded from a separate text file (using tab separated values format).

• Entered manually in the Ground Control dialog.

Advanced use

25

• Loaded from GPS EXIF tags (if present).

To load reference coordinates from a text file

1. Open Ground Control dialog using Ground Control... command from the Tools menu.

2. Select the Datum and Coordinate System used for reference coordinates data. For referencing in localEuclidean coordinates set Datum to Local coordinates.

3. Click Import button. Browse to the file containing recorded reference coordinates and click Openbutton.

4. The reference coordinates data will be loaded and the coordinate system will be initialized.

Note

• The tab separated format used for reference data is described in the next section.

To assign reference coordinates manually

1. Open Ground Control dialog using Ground Control... command from the Tools menu.

2. Select the Datum and Coordinate System used for reference coordinates data. For referencing in localEuclidean coordinates set Datum to Local coordinates.

3. In the Ground Control dialog box double click on x/y/z cells to assign values to corresponding co-ordinates.

4. Repeat for every marker/camera positions needed.

5. To remove unnecessary reference coordinates select corresponding items from the list and press Delkey.

6. Click Update button to apply changes and set coordinate system.

After assigning reference coordinates and initializing the coordinate system the referencing errors for eachreference point will be displayed in the Ground Control dialog and the largest error will be highlighted.At this point you can review the errors and decide whether additional refinement of marker locations isrequired (in case of marker based referencing), or if specific reference points should be excluded.

Note

• Unchecked reference points in the Ground Control dialog are not used for georeferencing.

• After adjusting marker locations on the photos, the coordinate system will not be updated auto-matically. It should be updated manually using Update button in the Ground control dialog box.

Reference coordinates file formatThe reference coordinates can be loaded into the Ground Control dialog using a simple tab separated textfile format. Each reference point is specified in this file on a separate line. Sample reference coordinatesfile is provided below:

# <label> <x> <y> <z>

Advanced use

26

IMG_0159.JPG 40.165011 48.103654 433.549477IMG_0160.JPG 40.165551 48.103654 434.724281IMG_0161.JPG 40.166096 48.103640 435.630558

Individual entries on each line should be separated with a tab character. All lines starting with # characterare treated as comments.

Records from the coordinate file are matched to the corresponding photos or markers based on the labelfield. Camera coordinates labels should match the file name of the corresponding photo including exten-sion. Marker coordinates labels should match the labels of the corresponding markers in the project file.All labels are case insensitive.

Note

• Tab-separated reference coordinates format doesn't include specification of the type of coordinatesystem used. The kind of coordinate system used should be selected in the Ground Control dialogseparately.

Performing measurementsPhotoScan supports measurement of distances between control points, as well as surface area and volumeof the reconstructed 3D model.

Distance measurementThe distance measurement tool enables measurements of direct distances between the points of the recon-structed 3D scene. The points used for distance measurement, as well as setting the reference distance,must be defined by placing markers in the corresponding locations. The reference distance or model co-ordinate system must be also initialized before the distance measurements can be performed.

For instructions on placing the markers, refining their positions and setting the reference system pleaserefer to the Setting coordinate system section of the manual.

To set reference distance

1. Place markers in the scene at the points with the known distance between them.

2.Select the Set Reference Distance tool using the corresponding toolbar button.

3. Click on the pair of markers with the known distance inbetween.

4. In the Set Reference Distance dialog box specify the distance between the points. Click OK buttonto set the reference distance.

The scale of the model will be determined and it will be possible to measure distance between other pointsof the model.

Note

• The initial distance displayed in the Set Reference Distance dialog box corresponds to the dis-tance calculated based on previously defined scale. Setting the new reference distance resets thepreviously defined model scale or coordinate system.

Advanced use

27

To measure distance

1. Place the markers in the scene at the locations to be used for distance measurement.

2.Select the Measure Distance tool using the corresponding toolbar button.

3. Click on the pair of markers to be used for distance measurements.

4. The distance between the selected markers will be displayed in the Measure Distance dialog box.

Note

• The distance values measured by PhotoScan are displayed in meters.

Surface area and volume measurementSurface area or volume measurements of the reconstructed 3D model can be performed only after the scaleor coordinate system of the scene is defined. For instructions on setting coordinate system please refer tothe Setting coordinate system section of the manual.

To measure surface area and volume

1. Select Measure Area and Volume... command from the Tools menu.

2. The model surface area and volume will be displayed in the Measure Area and Volume dialog box.Surface area is measured in square meters, while mesh volume is measured in cubic meters.

Measuring the volume can be performed only for the models with closed geometry. If there are any holesin the model surface PhotoScan will report zero volume. Existing holes in the mesh surface can be filledin using Close Holes... command from the Tools menu before performing volume measurements.

Note

• If the coordinate system is not defined for the model PhotoScan will allow to calculate area andthe volume of the model. However, the values obtained will be absolutely meaningless as theunit of measurement is undefined.

28

Appendix A. Graphical User InterfaceApplication Window

General view of application window.

Model viewModel view is used for displaying 3D data. The view of the model depends on the current state of process-ing and is also controlled by mode selection buttons on the PhotoScan toolbar.

Model may be shown in textured, solid, shaded or wireframe mode. Along with the model the results ofphoto alignment may be displayed. These include point cloud and camera positions for each photo.

PhotoScan supports the following tools for navigation in the 3D view:

Tool Keyboard modifier

Rotation Tool Default

Pan Tool Ctrl key pressed

Zooming Tool Shift key pressed

All navigation tools are accessible in the navigation mode only. To enter the navigation mode click the

Navigation toolbar button.

Graphical User Interface

29

Note

• Zooming into the model can be also controlled by the mouse wheel.

Photo viewPhoto view is used for displaying individual photos, as well as masks and markers they contain.

Photo view is visible only if any photo is opened. To open the photo double click on its name in theWorkspace or Photos pane.

Workspace paneOn the Workspace pane all elements available for the current project are diplayed. These elements mayinclude:

• List of chunks in the project.

• List of photographs in each chunk.

• List of markers in each chunk.

• 3D models in separate chunks.

Buttons located on the pane toolbar allow:

• Add chunks

• Add photographs

• Add markers

• Enable or disable certain photographs or chunks from processing in further stages.

• Remove items

Each element in the list is linked with the context menu providing quick access to some common opera-tions.

Photos panePhotos pane displays the list of photos of the active chunk in the form of thumbnails.

Console paneConsole pane is used for:

• Displaying auxiliary information

• Displaying error messages

• Python commands


30

Note

• To open Console pane select Console command from the View menu.

Menu Commands

File Menu

NewCreates an empty PhotoScan project.

Open...Opens PhotoScan project file.

Append...Appends existing PhotoScan project file to the cur-rent one.

SaveSaves PhotoScan project file.

Save As...Saves PhotoScan project file with a new name.

Export Model...Saves 3D model.

Export Points...Saves point cloud and/or estimated camera posi-tions.

Export Orthophoto...Generates orthophoto based on the reconstructedgeometry.

Export Dem...Generates digital elevation model based on the re-constructed geometry.

ExitQuits the application. Prompts to save activeproject.

Edit Menu

UndoUndo the last mesh editing operation.

RedoRedo the previously undone mesh editing opera-tion.

Delete FacesRemoves selected faces from the mesh.

Crop FacesCrops selected faces.

Invert SelectionInverts current selection.

Grow SelectionGrows current selection.

Shrink SelectionShrinks current selection.

Select Components...Selects isolated mesh components.

View Menu

Show PointsDisplays sparse point cloud, reconstructed duringphotos alignment.

ShadedDisplays 3D model in the shaded mode.


31

View Menu

SolidDisplays 3D model in the solid mode.

WireframeDisplays 3D model in the wire frame mode.

TexturedDisplays 3D model in the textured mode.

Show CamerasShows or hides camera positions, reconstructedduring image alignment.

Show TrackballShows or hides the trackball.

Show InfoShows or hides the mesh information on-screendisplay.

Perspective/OrthographicSwitch visualisation view between Perspective andOrthographic.

ConsoleShows or hides Console pane.

WorkspaceShows or hides Workspace pane.

PhotosShows or hides Photos pane.

Workflow Menu

Add Photos...Loads additional photos to be processed by Photo-Scan.

Align Photos...Generates camera positions and sparse point cloud.

Build Geometry...Generates 3D model geometry.

Build Texture...Generates 3D model texture.

Align ChunksAligns multiple chunks.

Merge ChunksMerges multiple chunks into the single model.

Batch Process...Opens Batch Process dialog box.

Tools Menu

Decimate Mesh...Decimates mesh to the target face count.

Close Holes...Closes holes in the model surface.

Import Mesh...Imports edited mesh from the external program.

Import Texture...Imports edited texture from the external program.

Export Texture...Exports model texture.

Display Mesh Statistics...Collects and displays mesh statistics.

Set Reference Distance...Sets model scale based on the reference distance.

Measure Distance...Distance measurement tool.


32

Tools Menu

Measure Area and Volume...Displays surface area and volume of the model.

Ground Control...Sets model coordinate system based on the key-points coordinates.

Import Cameras...Imports external camera orientation data.

Undistort Photos...Removes nonlinear distortions by warping sourcephotos.

View EXIF Data...Displayes EXIF data for the loaded photos.

Camera Calibration...Shows camera calibration dialog box.

Preferences...Shows preferences dialog box.

Run Script...Opens Python Run Script dialog box.

Photo Menu

MoveSelects navigation mode.

Rectangle SelectionRectangle selection tool.

Intelligent ScissorsIntelligent Scissors selection tool.

Intelligent PaintIntelligent Paint selection tool.

Magic WandMagic Wand selection tool.

Edit MarkersSwitches to marker editing model.

View PointsShows detected features used for alignment of thecurrent photo.

Add SelectionAdds current selection to the mask.

Subtract SelectionSubtracts current selection from the mask.


ResetResets mask for the current photo.

ImportCreates mask from alpha channel.

PropagateSets current mask for all photos.

ShadingTurns mask shading on or off.

Help Menu

ContentsDisplays help contents.

Check for Updates...Checks if PhotoScan update is available for down-load.

Register Product...Activates the product using the activation key.


33

Help Menu

About PhotoScan...Displays program information, version number andcopyright.

Toolbar ButtonsGeneral commands

NewCreates a new PhotoScan project file.

OpenOpens a PhotoScan project file.

SaveSaves a PhotoScan project file.

3D view commands

NavigationSwitches to the model navigation mode.

Rectangle SelectionRectangle face selection tool.

Circle SelectionCircle face selection tool.

Select VolumeVolume selection tool.

Create MarkerMarker placement tool.

Set Reference DistanceDefines reference distance.

Measure DistanceDistance measurement tool.

DeleteRemoves selected faces from the mesh.

CropCrops selected faces.

UndoUndo the last mesh editing operation.

RedoRedo the previously undone mesh editing opera-tion.

3D view settings

Show PointsShows/hides sparse point cloud, reconstructed dur-ing image alignment.

Show CamerasShows/hides camera positions, reconstructed dur-ing image alignment.

Show Aligned ChunksShows/hides aligned chunks.

TexturedDisplays 3D model in the textured mode.

ShadedDisplays 3D model in the shaded mode.

SolidDisplays 3D model in the solid mode.

WireframeDisplays 3D model in the wireframe mode.


34

Photo view commands

MoveNavigation tool.

Rectangle SelectionRectangle selection tool.

Intelligent ScissorsIntelligent scissors tool.

Intelligent PaintIntelligent paint tool.

Magic WandMagic wand tool.

Edit MarkersSwitches to marker editing mode.

View PointsShows/hides feature points used for alignment ofthe photo.

Add SelectionAdds current selection to the mask.

Subtract SelectionSubtracts current selection from the mask.


UndoUndo the last mask editing operation.

RedoRedo the previously undone mask editing opera-tion.

Rotate RightRotates the photo clockwise.

Rotate LeftRotates the photo counter clockwise.

Zoom InIncreases magnification.

Zoom OutDecreases magnification.

ShadingTurns mask shading on or off.

35

Appendix B. TroubleshootingPhoto alignment succeeds, but the resultingcamera positions appear to be wrong

The main reasons of the wrong photo alignment are listed below:

• Small overlap between the photos

• Possible object movement with respect to the background objects

• Insufficient amount of object surface details captured by the camera

• High level of blur or noise on the source photos

To obtain more information about the reason of alignment failure for the specific dataset it may be helpfulto inspect feature points used for alignment of the photos as seen by the PhotoScan algorithms.

To inspect feature points used for alignment

1. Open the photo to be inspected by double clicking its name in the Workspace pane. The photo willbe displayed in the main window.

2.Switch to the point view mode using View Points toolbar button. The points used for alignmentof the photo will be overlayed on the image.

Depending on the observed point placement, the following recommendations may help to solve the prob-lem:

Problem Possible reason Recommended solution

Not enough points wereused.

Low image quality ortoo small photo overlap.

Try making better photos with a better cameraplacement. Pay attention to the camera settings,like proper level of ISO. Use a tripod if required.Setting a proper lighting may be also helpful tocapture higher amount of surface details.

The majority of pointsused belong to the back-ground objects.

The object was not staticduring capturing or oc-cupies only a small areaon the photos.

Mask out background regions or consider shoot-ing another photo set so that the object spans sig-nificant area on the photos. If masking of the back-ground areas is used, make sure to check Constrainfeatures by mask option in the Align Photos dialogbox during processing.

Reconstructed geometry appears to be cut andsome important parts are missing

Usually this indicates that a wrong reconstruction volume was selected. By default PhotoScan uses anautomatic reconstruction volume selection algorithm, which may produce undesirable selections in somecases. All object parts outside of the selected reconstruction volume are cropped and are not included in

Troubleshooting

36

the final model. Too large reconstruction volume selections may be also undesirable as they result in largerprocessing time and memory consumption.

To overcome this problem a manual reconstruction volume selection tool should be used.

Note

• The photos must be aligned before the reconstruction volume can be defined.

To select the reconstruction volume manually

1.Select the manual reconstruction volume selection tool using Select Volume toolbar button.

2. Modify the displayed bounding box by dragging the corners of the box to the desired locations.

3. When reconstructing geometry make sure that Selection option is chosen from the Reconstructionvolume drop down list.

The photos included in the project file can'tbe opened and operations from the Workflowmenu fail

Probably the locations of photos have changed respectively to the location of the project file. The referencesto the original photos are stored in the project files in the form of relative paths. These references becomeinvalid when the project file is moved to another location alone, or when the photos are moved withoutchanging the project file placement.

To identify the expected photo location

1. Open the context menu of the photo by right-clicking its title in the Workspace pane.

2. Select Show Info... command from the context menu.

3. A dialog box with information on the selected photo including the path to the image data file willbe displayed.

To fix the problem move the photos to the original locations indicated by the data displayed in the infor-mation dialog box.

If the relative location of the project file in respect to the photos location must be changed, the project fileshould be resaved using the Save As... command from the File menu.

To move the project file in respect to the source photo locations

1. Open the project to be moved using Open... command from the File menu.

2. Select Save As... command from the File menu. Browse to the folder where the project file should beplaced and specify the destination file name for the project. Click Save to save the project.

After the project file was saved to the new location, the original project file can be removed, if necessary.

Documents

Pscan Pro En