AI Paper Atlas

Research map / Video / Video segmentation and tracking

Complementary Attention Gated Network for Pedestrian Trajectory Prediction

Jinghai Duan, Le Wang, Chengjiang Long, Sanping Zhou, Fang Zheng, Liushuai Shi, Gang Hua

AAAI 2022 · Video › Video segmentation and tracking · approach: Physics-inspired motion prediction · 54 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
Use complementary spatial and temporal attention with gated fusion and GMM endpointsPedestrian trajectory prediction · AVG (ETH and UCY)ADE: 0.34 -> 0.25Improved
Use complementary spatial and temporal attention with gated fusion and GMM endpointsPedestrian trajectory prediction · AVG (ETH and UCY)FDE: 0.58 -> 0.43Improved
Add complementary spatial attention instead of GCN (ablation)CAGN · AVG (ETH and UCY)ADE: 0.44 -> 0.25Improved
Add complementary spatial attention instead of SGCN (ablation)CAGN · AVG (ETH and UCY)ADE: 0.32 -> 0.25Improved
Add complementary spatial attention instead of Transformer (ablation)CAGN · AVG (ETH and UCY)ADE: 0.34 -> 0.25Improved
Add complementary temporal attention instead of LSTM (ablation)CAGN · AVG (ETH and UCY)ADE: 0.50 -> 0.25Improved
Add complementary temporal attention instead of Transformer (ablation)CAGN · AVG (ETH and UCY)ADE: 0.33 -> 0.25Improved
Add gated fusion instead of equal-weight fusion (ablation)CAGN · AVG (ETH and UCY)ADE: 0.27 -> 0.25Improved
Add gated fusion instead of equal-weight fusion (ablation)CAGN · AVG (ETH and UCY)FDE: 0.45 -> 0.43Improved
Use GMM endpoint distributions instead of a single Gaussian distribution (ablation)CAGN · AVG (ETH and UCY)ADE: 0.27 -> 0.25Improved
Use GMM endpoint distributions instead of a single Gaussian distribution (ablation)CAGN · AVG (ETH and UCY)FDE: 0.44 -> 0.43Improved

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