xvjiarui / VFS

Rethinking Self-Supervised Correspondence Learning: A Video Frame-level Similarity Perspective, in ICCV 2021 (Oral)

https://jerryxu.net/VFS/

xvjiarui/VFS Issues

Welcome update to OpenMMLab 2.0
Updated a year ago
Training Logs on Resnet18 Backbone
Updated 2 years ago
Failed to reproduce the result (finally solved, due to bugs in dataset setup)
Closed 2 years ago3
Pretrained model for the VIP dataset
Updated 2 years ago5
the pretrained model
Closed 2 years ago
Tuple object is not callable
Updated 2 years ago1
backbone.init_weights issue
Closed 2 years ago11
Kinetics400 videos size and training times
Closed 2 years ago1
How to run a demo to a new videos or bunch of frames?
Closed 2 years ago1
Testing on DAVIS gives error from rawframe_dataset.py
Closed 2 years ago4
Request the precomputed results on DAVIS17
Closed 2 years ago2
Issues relating to using ann_file_train for Kinetics400 dataset
Closed 2 years ago2
The split txt used for JHMDB and VIP
Closed 2 years ago5
Finding out frame similarity matrix between two video frames
Closed 2 years ago2
ModuleNotFoundError: No module named 'mmaction.version'
Closed 2 years ago2
Structuring Kinetics400 dataset
Closed 2 years ago2
How long does it take to train for 100 epochs?
Closed 3 years ago4
request for resnet pretrained on kinetics400
Closed 3 years ago4