atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Multimodal Interaction-Aware Trajectory Prediction in Crowded Space

AAAI 2020 · Probabilistic and scientific modeling › Molecules and proteins · approach: Multimodal trajectory prediction · 28 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 spatial interaction edges (Spatial-E) to the MDN temporal model (ablation)T-D1-1 · ETH/UCY averageADE/FDE (m): 0.48/1.11 -> 0.44/0.95Helped
replace L2 loss with mixture density output (ablation)T-1 · ETH/UCY averageADE/FDE (m): 0.48/1.11 -> 0.48/1.11No clear effect
sample 20 trajectories and take best instead of single mean prediction (ablation)ST-D1-1 · ETH/UCY averageADE/FDE (m): 0.44/0.95 -> 0.31/0.63Helped
sample from top-two mixture components instead of top one (ablation)ST-D1-20 · ETH/UCY averageADE/FDE (m): 0.31/0.63 -> 0.29/0.56Helped
sample from top-three components instead of top-two (ablation)ST-D2-20 · ETH/UCY averageADE/FDE (m): 0.29/0.56 -> 0.33/0.63Hurt
offset and velocity inputs with LSTM temporal model (T-1) vs plain LSTMLSTM · ETH/UCY averageADE/FDE (m): 0.69/1.35 -> 0.48/1.11Helped

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