AI Paper Atlas

Research map / Video / Action recognition

Pairwise Contrastive Learning Network for Action Quality Assessment

Mingzhe Li, Hong-Bo Zhang, Qing Lei, Zongwen Fan, Jinghua Liu, Ji-Xiang Du

ECCV 2022 · Video › Action recognition · approach: Action anticipation and latent actions · 37 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 PCLN branch (ablation)basic regression network · AQA-7average SRC: 0.8504 -> 0.8721Improved
add consistency constraint (ablation)basic regression + PCLN · AQA-7average SRC: 0.8721 -> 0.8795Improved
add PCLN branch (ablation)basic regression network · MTL-AQA FSPSRC: 0.8745 -> 0.8788Improved
add PCLN branch (ablation)basic regression network · MTL-AQA ESPSRC: 0.9095 -> 0.9196Improved
add consistency constraint (ablation)basic regression + PCLN · MTL-AQA FSPSRC: 0.8788 -> 0.8798Improved
add consistency constraint (ablation)basic regression + PCLN · MTL-AQA ESPSRC: 0.9196 -> 0.9230Improved
increase batch size from 8 to 32 (28 to 496 video pairs) (ablation)PCLN · AQA-7average SRC: 0.8276 -> 0.8795Improved
increase batch size from 32 to 64 (496 to 2016 pairs) (ablation)PCLN · AQA-7 DivingSRC: 0.8697 -> 0.8189Hurt
increase batch size from 8 to 64 (28 to 2016 pairs) (ablation)PCLN · MTL-AQA FSPSRC: 0.8729 -> 0.8798Improved
increase batch size from 8 to 64 (28 to 2016 pairs) (ablation)PCLN · MTL-AQA ESPSRC: 0.9094 -> 0.9230Improved
predict execution score then multiply by difficulty (ESP) instead of final score directly (FSP)PCLN · MTL-AQASRC: 0.8798 -> 0.9230Improved
use temporal encoder features instead of P3D 3DCNN features (ablation)basic regression network · AQA-7 diving_007predicted score stability: 83.07 fixed over 10 tests, ground truth 83.25Improved

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