AI Paper Atlas

Research map / Video / Action recognition

Anticipative Video Transformer

Rohit Girdhar, Kristen Grauman

ICCV 2021 · Video › Action recognition · approach: Action anticipation and latent actions · 240 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
add L_cls and L_feat anticipative losses (ablation)AVT with AVT-b backbone · EK55 val (RGB, anticipative training ablation)action top-5 accuracy: 25.9 -> 30.1Improved
add L_cls intermediate action loss (ablation)AVT with AVT-b backbone · EK55 val (RGB, anticipative training ablation)action top-5 accuracy: 25.9 -> 28.8Improved
add L_feat alone without L_cls (ablation)AVT with AVT-b backbone · EK55 val (RGB, anticipative training ablation)action top-5 accuracy: 25.9 -> 23.9Hurt
add L_feat to L_cls (ablation)AVT with AVT-b backbone · EK55 val (RGB, anticipative training ablation)action top-5 accuracy: 28.8 -> 30.1Improved
add L_cls and L_feat anticipative losses (ablation)AVT-h on IG65M irCSN152 features · EK55 val (RGB, anticipative training ablation)action top-5 accuracy: 31.4 -> 31.7No clear effect
train with anticipative losses, longer temporal context (ablation)AVT with AVT-b backbone · EK100 val (RGB)action recall@5: 11.2 -> 14.9Improved
train with naive loss, longer temporal context (ablation)AVT with AVT-b backbone · EK100 val (RGB)action recall@5: 11.0 -> 13.1Improved
replace TSN features with end-to-end AVT-b backbone (IN21+1k)AVT-h · EK100 val (RGB)action recall@5: 13.6 -> 14.4Improved
replace TSN features with irCSN152 IG65M featuresAVT-h · EK100 val (RGB)action recall@5: 13.6 -> 12.8Hurt
use IN21k instead of IN1k pretrainingAVT-b backbone · EK100 val (RGB)action recall@5: 13.4 -> 14.9Improved
replace RULSTM head with AVT-h on same TSN featuresAVT-h · EK100 val (RGB)action recall@5: 13.3 -> 13.6Improved
add object modality and late fuse with RGBAVT · EK100 valaction recall@5: 14.9 -> 15.9Improved

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