AI Paper Atlas

Research map / Video / Action recognition

Compound Prototype Matching for Few-Shot Action Recognition

Yifei Huang, Lijin Yang, Yoichi Sato

ECCV 2022 · Video › Action recognition · approach: Disentangled action recognition · 66 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 object features and multi-relation encoderCompound prototype matching · SSv2#, Kinetics, HMDB1-shot accuracy: 49.3 -> 59.6; 73.3 -> 81.0; 60.1 -> 80.3Improved
add object features by concatenation to baselines (ablation)MatchNet/ITA-Net/TRX · SSv2, Kinetics1-shot accuracy: 2~5 percent on SSv2, 2~6 percent on KineticsImproved
add multi-relation encoder (global-global only) without object features (ablation)ITA-Net, TRX, Ours · SSv2#, Kinetics1-shot accuracy: TRX 41.5 -> 44.4; 64.6 -> 68.9Improved
add multi-relation encoder together with object features (ablation)Ours · SSv2#1-shot accuracy: 49.1 -> 59.6Improved
add multi-relation encoder together with object features (ablation)TRX+ · SSv2#1-shot accuracy: 44.2 -> 48.4Improved
add global-object relation F_go to F_gg and F_oo (ablation)Ours · SSv2#1-shot accuracy: 57.7 -> 59.6Improved
add object-object relation F_oo to F_gg (ablation)Ours · SSv2#1-shot accuracy: 49.3 -> 57.7Improved
add global-global relation F_gg to F_go and F_oo (ablation)Ours · Kinetics1-shot accuracy: 77.1 -> 81.0Improved
add focused prototypes to global prototypes (ablation)Ours · SSv2#, Kinetics1-shot accuracy: 39.3 -> 59.6; 66.1 -> 81.0Improved
add global prototypes to focused prototypes (ablation)Ours · SSv2#, Kinetics1-shot accuracy: 44.0 -> 59.6; 76.7 -> 81.0Improved
increase prototype count with more focused than global prototypes (ablation)Ours · SSv2#, Kinetics1-shot accuracy: 59.6 -> 62.0; 81.0 -> 86.9Improved
replace multi-relation encoder with feature concatenation (ablation)Ours · UCF1-shot accuracy: 79.0 -> 87.7Improved

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