anuragranj / cc

Competitive Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentation
https://research.nvidia.com/publication/2018-05_Adversarial-Collaboration-Joint
MIT License
512 stars 63 forks source link

Runing test_flow on a real-world video #21

Open umarasif opened 4 years ago

umarasif commented 4 years ago

Hi,

Is there any demo script for producing optical flow on a test video? Specifically, I am interested in running test_flow.py on a video but the current code reads data (tgt_img, ref_imgs, intrinsics, intrinsics_inv) from a validation dataset loader. How do I extract these information from a test video?

Thanks

anuragranj commented 4 years ago

You would need to to write your own dataset loader. Here is a discussion on how to do it. https://discuss.pytorch.org/t/is-there-an-off-the-shelf-video-data-loader-to-use/3041