AI Paper Atlas

Research map / Video / Action recognition

VideoLT: Large-Scale Long-Tailed Video Recognition

Xing Zhang, Zuxuan Wu, Zejia Weng, Huazhu Fu, Jingjing Chen, Yu-Gang Jiang, Larry S. Davis

ICCV 2021 · Video › Action recognition · approach: Contrastive video representation learning · 52 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use running AP instead of constant beta=0.5 to set frames per class (ablation)FrameStack (ResNet-101, nonlinear) · VideoLTmAP (overall): 0.414 -> 0.532Improved
use running AP instead of class frequency to set frames per class (ablation)FrameStack (ResNet-101, nonlinear) · VideoLTmAP (overall): 0.520 -> 0.532Improved
use running AP instead of class frequency to set frames per class (ablation)FrameStack (ResNet-101, NetVLAD) · VideoLTmAP (overall): 0.663 -> 0.670Improved
use running AP instead of constant beta=0.5 (ablation)FrameStack (ResNet-101, NetVLAD) · VideoLTmAP (overall): 0.648 -> 0.670Improved
add FrameStack to the nonlinear model on ResNet-101 featuresnonlinear classifier · VideoLTmAP (overall): 0.516 -> 0.532Improved
add FrameStack to the nonlinear model on ResNet-101 featuresnonlinear classifier · VideoLT tail classesmAP (tail): 0.396 -> 0.417Improved
add FrameStack with NetVLAD on TSM featuresNetVLAD classifier · VideoLTmAP (overall): 0.660 -> 0.667Improved
add mixup in feature spacenonlinear classifier (ResNet-101) · VideoLTmAP (overall): 0.516 -> 0.495Hurt
add class-balanced samplingnonlinear classifier (ResNet-101) · VideoLTmAP (overall): 0.516 -> 0.507Hurt
add LDAM + DRWnonlinear classifier (ResNet-101) · VideoLTmAP (overall): 0.516 -> 0.518No clear effect
use TSM Kinetics-pretrained features instead of ImageNet ResNet-101 featuresFrameStack nonlinear · VideoLTmAP (overall): 53.2 -> 58.0Improved
use focal loss instead of BCE in FrameStack (ablation)FrameStack (ResNet-101, nonlinear) · VideoLTmAP (overall): 0.521 -> 0.532Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy