seg2vid
seg2vid copied to clipboard
Video Generation from Single Semantic Label Map
Hi, Both UCF and KTH datasets require **datalist** I'm not sure how to generate those **datalists** Can you provide any hint or explanation?
https://github.com/STVIR/seg2vid/blob/89f5ca98dcac4f0fd9e592302b7a1430918d6779/src/dataset.py#L4 function uses `CITYSCAPES_VAL_DATA_PATH`. Looks like a typo.
Can you please tell how much time you trained your network?
Hi, could you provide the script that you have used for the segmentation of the video training set?
Hi: in the dataset_path.py, there is no path for the segment masks on the ucf and kth datasets. I wander if i can only use the test_refine.py to test the...
Hi: I followed your instructions in readme,but I got the following error: Traceback (most recent call last): File "test_refine_w_mask_two_path.py", line 149, in a.test() File "test_refine_w_mask_two_path.py", line 124, in test y_pred_before_refine,...
Hi, thanks for the nice work ! I have a question about the models you share. Did you train separate models for each category in ucf101 and kth ?
Hi, Thanks so much for releasing the code! Unfortunately, the code seems untested, as there are many typo's which raise runtime errors (e.g. `uitls` and `datset` throw exceptions). Could you...