AI Paper Atlas

Research map / Video / Action recognition

What and How Well You Performed? A Multitask Learning Approach to Action Quality Assessment

Paritosh Parmar, Brendan Tran Morris

CVPR 2019 · Video › Action recognition · approach: Disentangled action recognition · 210 citations (OpenAlex)

Check 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 action classification and commentary generation as auxiliary tasksC3D-AVG AQA on MTL-AQA · MTL-AQASpearman's rank correlation: 89.60 -> 90.44Improved
Add action classification and commentary generation as auxiliary tasksMSCADC AQA on MTL-AQA · MTL-AQASpearman's rank correlation: 84.72 -> 86.12Improved
Add action classification without commentary generation (ablation)C3D-AVG AQA · MTL-AQASpearman's rank correlation: 89.60 -> 89.62No clear effect
Add commentary generation without action classification (ablation)C3D-AVG AQA · MTL-AQASpearman's rank correlation: 89.60 -> 88.78Hurt
Combine action classification with commentary generation (ablation)C3D-AVG AQA · MTL-AQASpearman's rank correlation: 88.78 -> 90.44Improved
Add action classification without commentary generation (ablation)MSCADC AQA · MTL-AQASpearman's rank correlation: 84.72 -> 85.76Improved
Add commentary generation without action classification (ablation)MSCADC AQA · MTL-AQASpearman's rank correlation: 84.72 -> 85.47Improved
Add action classification and commentary generationC3D-AVG AQA · UNLV DiveSpearman's rank correlation: 83.83 -> 88.08Improved
Add action classification and commentary generationMSCADC AQA · UNLV DiveSpearman's rank correlation: 79.79 -> 80.60Improved
Add action classification and commentary generationC3D-AVG AQA · MTL-AQA with reduced training dataSpearman's rank correlation: 77.27 -> 83.52; 69.63 -> 72.09; 64.17 -> 68.16Improved
Train AQA-oriented representations jointly with action and commentary supervision, rather than using action-recognition featuresC3D-AVG-MTL representations · MTL-AQA convolutional-layer featureslinear-regressor Spearman correlation: 74.26 vs. 71.01; 77.95 vs. 71.39; 82.78 vs. 73.13; 86.18 vs. 76.34; 85.75 vs. 73.69Improved

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