AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Sparse Instance Conditioned Multimodal Trajectory Prediction

Yonghao Dong, Le Wang, Sanping Zhou, Gang Hua

ICCV 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Multimodal trajectory prediction · 40 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 sparse instance, memory refinement and clustering to the baseline (ablation)SICNet · ETH-UCY (avg)ADE/FDE: 0.46/0.94 -> 0.19/0.33Improved
add clustering post-process on top of sparse instance and refinement (ablation)SICNet · ETH-UCY (avg)ADE/FDE: 0.22/0.39 -> 0.19/0.33Improved
use mask threshold 0.2 instead of 0 or 1 (ablation)SICNet · ETH-UCY (avg)ADE: 0.21 (mu=0) -> 0.19 (mu=0.2)Improved
use sparse instance instead of goal or full instance as the condition (ablation)SICNet · ETH-UCY (avg)ADE/FDEImproved
use SICNet instead of MemoNet with small best-of-KSICNet · SDD (K=10)ADE/FDE: 11.26/19.64 -> 10.27/18.13Improved
SICNet vs MemoNet on SDD final displacement error at K=20SICNet · SDD (best-of-20)FDE: 12.66 -> 13.65Hurt
SICNet vs PCCSNet at K=5 on SDD ADESICNet · SDD (K=5)ADE: 12.57 -> 12.82Hurt

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