Closed jkamalu closed 2 years ago
Thank you for your interests in our work!
Unfortunately, we don't have person id information in our current TSV files. So, there is probably no way to find the correspondence between the person id and the timestep, especially for 3DPW :(
For H3.6M, I think it may be possible to do it.
H3.6M has only a single person in each video. We just need to find timestamp. Timestamp can be estimated based on the image key:
For example, given an image key images/S1_Directions_1.54138969_000001.jpg
,
the format is images/[subjectID]_[Action]_[CameraID]_[FrameID].jpg
Thank you for your interests in our work!
Unfortunately, we don't have person id information in our current TSV files. So, there is probably no way to find the correspondence between the person id and the timestep, especially for 3DPW :(
For H3.6M, I think it may be possible to do it.
H3.6M has only a single person in each video. We just need to find timestamp. Timestamp can be estimated based on the image key: For example, given an image key
images/S1_Directions_1.54138969_000001.jpg
, the format isimages/[subjectID]_[Action]_[CameraID]_[FrameID].jpg
Hi Kevin, why the CameraID is "1.54138969", a float value. In H36M cam_idx is a int value.
Hello,
Thank you for publishing such a complete project for your work. It is very much appreciated!
Is there a way to figure out the correspondence between the person id and the timestep for video datasets? For example, I want to edit the dataloader to randomly sample image crops of the same person in chronological order e.g.
[person i @ frame 0 in video k, person i @ frame 1 in video k, person i @ frame 2 in video k, etc. ]
where i and k are randomly sampled.Is this possible given the meta data for 3DPW, Human 3.6 Million, or any of the other video based datasets? Or would I have to generate my own .tsv files? I would appreciate any and all help you can give!
Thanks