ScanNet
ScannerApp with Bundlefusion
Hello, I am going to use color and depth info data from ScannerApp to reconstruct 3d model with Bundlefusion. Now I can get data from ScannerApp(.camera, .depth, .h264, .imu, .txt), but How can I merge these files to one file(.sens)? Thank you!
Have you managed to do it @CDElite ? I'm interested as well to make it work from .pgm .ppm and .dump from NYU dataset.
At this page there's more information on the Sens file, but I don't understand how to make the conversion to Sens. https://github.com/ScanNet/ScanNet/tree/master/SensReader
@franmts i wrote a simple program which make the conversion to Sens. and now i can run the bundlefusion with that .sens but the result is not so ideal because of the lack of camera position
@CDElite
For the camera positions, I think that's the job of bundlefusion, which runs global bundle adjustment at almost every frame. Theoretically it doesn't need any help on providing camera positions.
Would you mind sharing your .sens system as well for me to try it out?
@franmts But the .sens file includes camera position for every frame in it's structure. Do you have any idea for camera position? It's all right to share my code. I can send e-mail to you if you provide your email address.
Couldn't find the .sens file structure.great that you figured out on your own.
I posted ".sens and NYU reconstructions" thread directly on Bundlefusion GitHub, but so far didn't get any support answers.
If you can send to [email protected] would save a lot of iteration time. Thanks
@CDElite - It would be great to get the sens creation script as well ([email protected]). Thank you! Guy
@franmts How is it going on? Did you figure out the Depth To Color Calibration?
@CDElite, thanks for the sens system. Though, we realized bundlefusion is not the ideal solution for us. We're looking into alternatives, like ARKit and faster photogrammetry. Did you advance on the reconstruction quality?
@CDElite Hi, I am also trying to use bundle fusion with this app. Is that possible you can send me the code to create .sens file. Thank you very much. My email is [email protected]
@CDElite Hi, I'm also trying to reconstruct 3d model with Bundlefusion. If it's convenient for you, could you please send me the sens creation script as well?([email protected])Thanks sincerely for your kind help! Best!
@CDElite Hi, can you send me the code to create .sens file. Thank you very much. My email is [email protected]
@CDElite @franmts Hi, if possible, could u guys also send me ([email protected]) the code for creating the .sens files. Great thks
@CDElite @franmts Hello, could you please send me ([email protected]) the code for creating .sens files if convenient? I got stuck in this for quite a while... Thank you so much!
@franmts How is it going on? Did you figure out the Depth To Color Calibration?
@franmts i wrote a simple program which make the conversion to Sens. and now i can run the bundlefusion with that .sens but the result is not so ideal because of the lack of camera position
Hi CDElite! I wrote a program to converse to Sens. , too. But when I take the conversed .sens file as input of BundleFusion, it doesn't work. You can see the error in error Could you please give me some addvice?
@CDElite Hi! I would also be interested in having your conversion code if you don't mind sending it. Thanks! [email protected]
@CDElite Hello, if you possible, could you send me the code to create the .sens file too. I would be grateful if you would help me. Thank you very much! My email address is [email protected]
Hi all, I found that they have made a python script to read from .sens files. Check it out here: https://github.com/ScanNet/ScanNet/tree/master/SensReader/python
This does not come with a way to create .sens files from image data, however, so I created a python script that does that. Feel free to send me an email at [email protected] if you want it @Hi610