AI Paper Atlas

Research map / 3D vision / 3D generation

SiT Dataset: Socially Interactive Pedestrian Trajectory Dataset for Social Navigation Robots

Jongwook Bae, Jungho Kim, Junyong Yun, Changwon Kang, Jeongseon Choi, Chanhyeok Kim, Junho Lee, Jungwook Choi, Jun Won Choi

NeurIPS 2023 · 3D vision › 3D generation · approach: Scene-aware human motion synthesis · 3 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 semantic map scene encodingY-Net · SiT trajectory predictionADE5: 1.527 -> 1.361Improved
add semantic map scene encodingY-Net · SiT trajectory predictionADE20: 0.836 -> 0.675Improved
add semantic map scene encodingY-Net · SiT trajectory predictionFDE20: 1.878 -> 1.547Improved
add semantic map scene encodingNSP-SFM · SiT trajectory predictionADE5: 1.346 -> 1.061Improved
add semantic map scene encodingNSP-SFM · SiT trajectory predictionFDE20: 1.087 -> 0.925Improved
fuse camera with LiDAR instead of camera only3D detectors · SiT 3D pedestrian detectionmAP: 0.244 -> 0.531Improved
use SECOND voxel backbone instead of PointPillars backboneCenterPoint · SiT 3D pedestrian detectionmAP: 0.414 -> 0.518Improved
use SECOND voxel backbone instead of PointPillars backboneTransFusion · SiT 3D pedestrian detectionmAP: 0.390 -> 0.531Improved
use SECOND voxel backbone instead of PointPillars backboneFutureDet · SiT end-to-end motion predictionmAPf: 0.037 -> 0.053Improved
use CenterPoint detector instead of PointPillars with AB3DMOTAB3DMOT · SiT 3D pedestrian trackingsAMOTA: 0.4110 -> 0.4841Improved
use CenterPoint tracker instead of AB3DMOTCenterPoint detector · SiT 3D pedestrian trackingsAMOTA / IDS: 0.4841 -> 0.6070 sAMOTA; 554 -> 1136 IDSMixed
use camera+LiDAR TransFusion-V instead of LiDAR CenterPoint-Vdetector · SiT 3D pedestrian detectionAP(0.25): 0.397 -> 0.318Hurt

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