Open Lijin002 opened 2 years ago
Hi, I encountered the same problem as you. My ./rio_renderer_render_all
and ./align_poses
worked in the two example data, but didn't work in the whole dataset of 3RScan (-v2). I can't figure it out. Have you solved this problem?
Hi, I've solved the problem by using the right version -v2. Is there any Error showed in your terminal?
Thanks for replying. I ran ./rio_renderer_render_all
following 3DSSG data_processing, it shows:
xxx@xxx:~/3dssg/3RScan/c++/rio_renderer/build$ ./rio_renderer_render_all /home/xxx/3dssg/3DSSG-wu/data/3RScan 754e884c-ea24-2175-8b34-cead19d4198d sequence 2
/home/xxx/3dssg/3DSSG-wu/data/3RScan/754e884c-ea24-2175-8b34-cead19d4198d/sequence//_info.txt
/home/xxx/3dssg/3DSSG-wu/data/3RScan/754e884c-ea24-2175-8b34-cead19d4198d/sequence//_info.txt
已放弃 (核心已转储)
And the ./align_poses
shows the following:
xxx@xxx:~/3dssg/3RScan/c++/rio_lib/build/bin$ ./align_poses /home/xxx/3dssg/3DSSG-wu/data/3RScan 754e884c-ea24-2175-8b34-cead19d4198d sequence
loading /home/xxx/3dssg/3DSSG-wu/data/3RScan/3RScan.json
Error reading /home/xxx/3dssg/3DSSG-wu/data/3RScan/3RScan.json unexpected end of input
saved 0 poses, in /home/xxx/3dssg/3DSSG-wu/data/3RScan/754e884c-ea24-2175-8b34-cead19d4198d/sequence
I downloaded the 3RScan-v2 following the project link with commands python download.py -o /path/to/3RScan/ --type semseg.v2.json
and python download.py -o /path/to/3RScan/ --type labels.instances.annotated.v2.ply
.
The sample of my downloaded dataset shows below.
Have I done anything wrong?
I've encountered this error before, but there are many reasons for it, so I can't pinpoint what caused this either. Judging by the error message, I wonder whether you have downloaded the sequence.zip of each scene in 3RScan. The 3RScan.json is also not so exactly right, maybe check the file 3RScan.json or redownload it? ps. the link to download sequence.zip https://github.com/WaldJohannaU/3RScan
Got it. I need to unzip all the downloaded sequence.zip
. I am working on it, thanks a lot.
Hi, thank you so much for your great work and the 3RScan dataset! Recently, I've encountered some errors when I tried to render all camera poses at once for one scan in 3RScan. I've compiled the rio_render project under guidance on Github, but there is no output in terminal, and no rendered results, showed as following pictures.
I tried with the example data and succeed, as follows, there are files named xxx.rendered.xxx.
I wonder whether the version of 3RScan dataset is the reason. I tried with 3RScan-v1, instead of v2, because the download link of 3RScan-v2 leads to 404 page. One of scenes I tried contains the following files.
Could you please help me figure out what's wrong and how to solve the problem? Thank you for your time and looking forward to your reply!