anuragranj/cc
Competitive Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentation
PythonMIT
Issues
- 3
Link of Pretrained Models has been broken.
#30 opened by pureYioYi - 2
Code does not reproduce results from Figure 4
#28 opened by mmbannert - 7
scipy.misc.imresize removed in scipy 1.3.0.
#14 opened by cgo - 3
viz3 = np.vstack((255*tgt_img_viz, 255*depth_viz, 255*mask_viz... in test_mask
#17 opened by gagajian - 8
- 0
- 3
- 15
Training script
#5 opened by charliememory - 0
about log-terminal
#26 opened by jilner - 11
About the output of the posenet
#7 opened by wangy9711 - 1
About correlation operation。
#25 opened by wenmingMeng - 1
Image size used in training and testing
#24 opened by JohnnieXDU - 0
Large EPE for optical flow evaluation
#23 opened by DeckerDai - 0
Regarding training loop.
#22 opened by ezorfa - 1
Runing test_flow on a real-world video
#21 opened by umarasif - 0
Regarding cityscapes dataset.
#20 opened by ezorfa - 8
Unable to train Dispnet and Posenet.
#19 opened by ezorfa - 11
About Rigidity masks
#18 opened by ezorfa - 0
- 15
spatial_correlation_sampler
#8 opened by babahaji - 1
--spatial-normalize
#11 opened by guangmingw - 1
Which weights to load
#12 opened by wenxin-bupt - 2
flownetC6 pretrained model
#13 opened by sunnyHelen - 1
run_inference.py pretrained weight
#9 opened by babahaji - 3
About GPU memory
#6 opened by zmlshiwo - 3
Details about the training algorithm
#4 opened by jmendozais - 2
When are u planning to release the code?
#1 opened by chowkamlee81 - 2
Is "Unsupervised Learning of Multi-Frame Optical Flow with Occlusions" code available ?
#2 opened by pinkfloyd06 - 0
Will this code still be available soon?
#3 opened by C2H5OHlife