Jump to content


  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by cdschroer

  1. I'm with leszekp on this. The software expects that nothing changes except the light positions. The images from the lower angles should be darker. Don't compensate with the lights, and don't exposure comp those images. You might want to read through the guide to highlight image capture on the CHI website - even though you are building a dome - because some core principles like this are described there. Carla
  2. Well, you can get decent visualizations if you violate this rule, but you won't have good normals, as you move farther away from the center of the subject. If you use normals visualization or specular enhancement with no color (both rely heavily on the normals data) you will see problems around the edges. So, it's all about "fit for purpose". If you want good normals for the whole subject, then you need the standoff. If you just want some relighting to get a decent screen shot, you might get away with less.
  3. Dear weiwei - do not worry. You set up a new topic, which was appropriate, and there is some information here that hopefully is useful. Also others should be able to find this, and I deleted the duplicate. Sometimes it is helpful for me to remind people, since this does happen from time to time. Welcome to the forums! Carla
  4. The recommended standoff of the lights for RTI is 2 -4 times the diagonal size of the subject. This is described in the Guide to Highlight Image Capture on page 23 (see below) If you use less than 2 times the size of the subject, you will not get accurate surface normals for the area outside the center of the subject that doesn't meet the light standoff requirement. I'm asking one of the computer graphics folks to reply as to the technical reasons for this. I know from experience and testing that this is true. Carla PS: I removed this identical question which you added to the end of another unrelated topic in the photogrammetry forum. This is appropriately a separate topic, and the discussion should take place here. As a reminder to folks - it is not helpful to post the same question in more than one place on the forums. Please find the most appropriate forum, and either add to an existing topic (if your question is part of that topic) or start a new topic. Thanks!
  5. Katie - It looks like you gave RTIBuilder the path to the ptmviewer instead of the ptmfitter. You need to change it to the ptmfitter - which you also download form HPLabs. It may have been in the same package you already downloaded. In that case, just open your existing RTIBuilder file for your project (instructions towards the end of the User Guide - for "open existing project) and when you get the final screen where you can "execute" building your RTI or PTM, change the path to the ptmfitter there.'' Carla
  6. Thanks Dave - yes - we try to keep to more general photogrammetry issues here in this space because Agisoft runs forums for PhotoScan. We are happy to entertain general questions about setup, good practice, equipment, photogrammetric approaches, etc. That said - there are some odd behaviors within photoscan related to masks whether you are doing an individual image, or a whole set. Crystal is correct that you need to select the folder containing the images, and the background image. If that folder is already selected then I usually select another folder, then the one I want to use before saying OK. Go to the Agisoft forums if you continue to have difficulties - and make sure you include what version of photoscan you are using, and what platform you are on (Mac, windows, linux - and OS level) Carla
  7. Hernandez - I'm not clear what problem you are having. First you need to have a .ptm or a .rti file. If instead you have a file of some other type (like Fmck showed) then this isn't a valid rti file. My guess is this is the result of something like the WedRTIViewer which does pre-processing on an rti or ptm file to set it up to work with a web based viewer. If that is the case - then the resulting folder has to be set up on a webserver with the appropraite HTML code to call it. This is described on the web page for the WebRTIViewer. I'm not an expert on that. If you want to use the regular RTiViewer - then you need an rti or ptm file. Carla
  8. Loa - We wish we could include the PTMfitter software, but this license issue (from HP) prevents it. The easiest way to check for spaces in your pathname is to go to the "assembly-files" folder created by RTIBuilder and open one of the .lp files (it is just a text file - you can open it in any text editor) There you will see the full path to the image files. You will be able to see if you have any spaces. If you do - move the whole folder (with the .xml log file, and the jpeg-exports folder, etc) to a correct location - or rename the needed directories. Then you should be able to build an RTI from the existing project without having to detect the spheres and highlights again. How to do that is described towards the end of the user guide - and also elsewhere in the forums. If the filenames themselves have spaces, then you will have to start over after renaming them. Good luck!
  9. Khalid - I can get both files from these links. The RTI spec was uploaded to this forum - so you have to be logged in to the forum to get it. I should have specified that in my earlier post. What happens if you click on the link? (For me, it downloads the file to my downloads folder) Carla
  10. Dear Loa - Sorry you have had such difficulty getting the software installed. It is very well documented in the User Guide that the PTMFitter is not included with the software and must be downloaded and installed from HP's website. This is because the RTIBuilder and HSHFitter software are both available under an open source license (GNU GPLv3) and the ptmfitter has a free for non-commerical use license with no source code. That license from HP does not allow us to bundle the software together, though it does allow you to download it for free and tell RTIBuilder where you installed it. You should be able to run the hshfitter without installing the ptmfitter. The "unknown error" message is likely not an installation problem. It is described in multiple places on these forums. The highest likelihood is that you have spaces somewhere in your image filenames and/or pathnames (i.e. the directories the files are stored in - all of them up to the root directory on whatever drive they are stored in) This is also a well documented issue in the user guide as well as the forums. The User guide should be where you turn to first to understand how to work with the software. It is free software and we have worked hard to provide good documentation, as well as these forums. The software is quite old at this point, and we would like to find funding to do an update. In the meantime - the software does work, albeit with a few quirks. Carla
  11. You need to give RTIBuilder the folder that contains the jpeg-exports folder - not the jpeg-exports folder itself. In other words, choose the folder one level up from the jpeg-exports folder. This is documented in the user guide, which can be found here: http://culturalheritageimaging.org/What_We_Offer/Downloads/Process/index.html Carla
  12. IT looks like only one of these is a ptm file - the first one. The other are all a different (.dat) format. The PTM file at 1KB is not a valid file. The file sizes would always be larger than that. I'm not sure how these were shared with you. Is it possible that these files are the result of some pre-processsing step for Web distribution? There are a couple of web viewers out there that require that you take a PTM or RTI file and run it through a program to prepare it for Web distribution. If that's the case, then you can only open the result with the web viewer, after the files are properly staged on a web server. What you want is just the regular full size RTI (or ptm) file. Carla
  13. Thanks for the additional details. I'll note that you can associate a photo with a group of RTI image sets - which doesn't totally get what you are asking for, but is simpler than associating with each image set. Also, the ability to "duplicate" image sets and/or groups of image sets will also automatically associate the photo. I agree that these ideas are "workarounds" for what you are asking for, but thought I would throw them out there in the hope that you might find the suggestions useful. Carla
  14. Hey guys! Thanks for figuring this out. I'm on the road doing training so just saw this. We definitely should make the default be to install the database. This probably needs its own topic where we can show the details and folks can find it easily. When I'm back on a computer (rather than my phone) I'll get that set up. Note that you may may need to start Postgres after rebooting your computer- or you will see the same message.
  15. Dave - thanks for the feedback. We are planning to internationalize the software to make supporting different localizations easier. I agree that dates can be tricky - and maybe we should just make the month and date formats clear as a start. Thanks for bringing up the dome scenario. Most folks are doing the highlight method of RTI - and this is a little bit skewed that way. Remember that you can enter as much or as little data as you want - so you can easily skip the string length - or use it to record the radius (i.e. the light standoff on your dome). You can make your own equipment category for a dome. Put in whatever you want about it, and associate an image or two of the dome with any groups of image sets - or image sets you wish. Good point about operator roles. Initially we were thinking about collecting the initial event of capturing the image sets, and so you can make these whatever level of specificity you want. I use "image capture" for a person playing multiple roles in the image capture. Good point about making it possible to associate the same person as the image processor. Thanks for the feedback! Carla
  16. Thanks for the feedback Dave. Here are a couple of notes: 1. You can run inspector without the database from DLNCC - and that might be the easiest way to use it when you are just getting started. This is also great for checking your own legacy data, that isn't in the DLNCC tool. If you want to do this you can turn off using the DLN in the "options" pane. You can turn it back on any time you like. 2. If you want to use it with the sample data, you will need to change the path for the sample data to wherever you installed it on your hard drive. I should have included that in the README for the sample data - sorry about that. If you use Inspector with DLNCC - it assumes it can find the files and folders in the stored path for the project. So, update the path at the project level, and keep the sample files under that and it should work. Thanks for taking the time to try out the tools! I think you will find the Inspector reports really useful, once you get over these early hiccups. Carla
  17. Welcome to the forum for the DLN:Inspector tool - one of the first two tools in the Digital Lab Notebook (DLN), a software pipeline made up of open source software and associated good practices. In this forum, we encourage you to share your questions, issues, or comments about using DLN:Inspector. The open source DLN:Inspector tool can be used to validate image sets created for Photogrammetry or Reflectance Transformation Imaging (RTI) You can download both the DLN tools from this page: http://culturalheritageimaging.org/What_We_Offer/Downloads/DLN/index.html There are Instructional videos describing how to work with the tools on this Vimeo channel: https://vimeo.com/channels/digitallabnotebook
  18. Welcome to the forum for the DLN:CaptureContext (DLNCC) tool! DLNCC is one of the first two tools in the Digital Lab Notebook (DLN), a software pipeline made up of open source software and associated good practices. In this forum, we encourage you to share your questions, issues, or comments about using DLNCC. DLNCC is all about collecting rich metadata for your photogrammetry and Reflectance Transformation Imaging (RTI) projects. You can easily reuse data about equipment, people, locations, subjects and more. The data can be exported as Linked Open Data mapped to the CIDOC Conceptual Reference Model (CRM) for archiving and reuse. And, you don't need to know anything about these metadata standards to produce these results. You can download the Capture Context tool, along with a user guide from this page http://culturalheritageimaging.org/What_We_Offer/Downloads/DLN/dln-cc/index.html There are Instructional videos describing how to work with the tools on this Vimeo channel: https://vimeo.com/channels/digitallabnotebook Please give us your feedback!
  19. Sorry to be late in responding - the issue when you see the message "The jpeg-exports folder is missing in your project directory" is that you didn't give RTIBuilder the correct folder. When starting a new project, it is looking for a project level folder, that includes a jpeg-exports folder where the images are to be found. The folder MUST have that name, and you have to give it a folder one level up from that. This is all documented in the user guide. Carla
  20. That functionality is not in the RTIViewer. It always starts at high noon in the default mode. Though it easy to see other views using the bookmarks. Carla
  21. The videos are up and running again! https://www.metmuseum.org/about-the-met/conservation-and-scientific-research/projects/rti-symposium/
  22. 2-4 times is the minimum distance - you can always work from farther away - as long as you have good even coverage across your subject from the light distance. For example, we do this under microscopes, and we have the light much farther away, because we are imaging small things. So, you want a comfortable working distance that is easy for you to set the distance. For small things it is easy for one person to hold the string, and the flash. We often set the camera on a timer and then just move the light, and have it take images. There are several ways to use a timer, the simplest is being tethered to a computer. For example, Canon cameras come with free software - the EOS Utility that allows you to automatically take pictures every X number of seconds. Also this way the images can be named and numbered and saved to an appropriate location in your computer. Carla
  23. The first thing I recommend is reading through the guide to highlight image capture - a free download from the CHI website: http://culturalheritageimaging.org/What_We_Offer/Downloads/Capture/index.html Some of your questions are answered there. The sphere size is based on the resolution of your camera, and the field of view you have set up with the lens you are using. The recommendation (found in the guide) is to have a minimum of 200 pixels across the diameter of the sphere. The key to avoiding shadows is sphere placement. There are some good examples in the guide. The number of images we recommend is 36-48. This is not a function of the size of the subject, but about collecting a good set of data to aide the software in accurately calculating the reflectance properties and surface normals. Carla
  24. Hmm - there should be an .lp file once you get to the execute stage. Can you check the .xml log file? It should be at the top level under the project directory. It should have the path, and also other information, and might give some clue as to the problem. You can attach it to a message here or email it to me: info at c-h-i.org (I'm trying to avoid the email address being detected by bots - but I think you can figure out what I mean. This seems to be a very unusual case. Carla
  25. The screenshot you attached and that you linked to, both show the contents of the jpeg-exports folder. You need to look inside the assembly-files folder for a file called .lp. This is the light position file. There may be more than one of them, depending on whether you cropped your images. It doesn't matter which one you look at. Carla
  • Create New...