TSAM
TSAM copied to clipboard
MaureenZOU
Hi author, please see the code in line 27: model_urls = { 'resnet50': '../libs/model_weights/TSM_imagenet_resent50.pth', 'resnet50_gated': '../libs/model_weights/TSM_imagenet_resent50.pth', } Why resnet50 and resnet50_gated point to the same place?
Hi author, there are too many download links from https://davischallenge.org Can you please point out which one do we use?
Hello Zou, Thanks for your open-source codes, I have a question on metric calculation. In the `_evaluate_test_videos` function (from line 285 to line 294) in `inference.py`, you transform all the...
I would like to know how to train with object masks since the data config are all about random masks Thanks for your great work!
Hi, Xueyan Here is Figure 2 (2) of your paper.  The inputs of the two conv layers are different: one is the shifted features and the other one...
Hi, Thanks for releasing code! I have a question about optical flow. Why you use "-flow" instead of "flow" in the warping function? 
I have found two mask folds in your readme, i think they are for test? but for FVI dataset, i only find one kind mask? can you provide the other...
Dear Xueyan, May I ask how do you calculate the Flow Validity Mask that you described in Section 3.3? Is this part of code in src/model/tsm/temperal_shift.py? But there is no...
Many thanks for publishing your code. May I ask is it possible to also provide the data including the flow and flow mask used in your training time? Many thanks!