AI Paper Atlas

Research map / Video / Action recognition

Generative Action Description Prompts for Skeleton-based Action Recognition

Wangmeng Xiang, Chao Li, Yuxuan Zhou, Biao Wang, Lei Zhang

ICCV 2023 · Video › Action recognition · approach: Skeleton-based action recognition · 142 citations (OpenAlex)

Read the paper

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 generative action-description prompts (GAP) to CTR-GCNCTR-GCN · NTU RGB+D 120 X-Subaccuracy (%): 88.9 -> 89.9Improved
add GAP to ST-GCN backbone (ablation)ST-GCN · NTU120 X-Sub jointaccuracy (%): 82.6 -> 83.8Improved
use label name as text prompt (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 84.6 -> 84.8No clear effect
use synonym or paragraph prompts (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 84.6 -> 85.2Improved
use body part description prompts (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 84.6 -> 85.4Improved
combine synonym and body part descriptions (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 85.4 -> 85.5Improved
use more body parts and multi-part contrastive loss (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 85.2 -> 85.4Improved
increase partition from four to six parts (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 85.4 -> 85.4No clear effect
use GPT-3 generated descriptions instead of manual or HAKE descriptions (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 85.2 -> 85.5 (manual), 85.3 -> 85.5 (HAKE)Improved
use generated prompts instead of learnable prompt vectors (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 85.2 -> 85.5Improved
swap CLIP text encoder for BERT (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 85.2 -> 85.2No clear effect
use JSD loss instead of KL divergence contrastive loss (ablation)GAP · NTU120 X-Sub jointaccuracy (%): 85.5 -> 85.7Improved

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