google-deepmind / google-deepmind/kinetics-i3d

LiteFlowNet over OpenCV TV-L1 optical flow algorithm. Can we ?

Open
#41 5 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
1.8k
Forks
467
PR merge metrics
No merged PRs in 30d

Description

I have read about other's people problematic experiences with OpenCV's TV-L1 optical flow algorithm about how time consuming it is and I have also witnessed it myself first-hand.
My question is simple, is it legitimate to choose another Optical Flow Estimation way, for instance the LiteFlowNet (http://mmlab.ie.cuhk.edu.hk/projects/LiteFlowNet/) which is currently the state-of-the-art at CVPR 2018. will this affect the results, especially when I intend to use the Flow Kinetics-i3d model solely for feature extraction purposes ?
Thanks in advance.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.