AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

Hetecooper: Feature Collaboration Graph for Heterogeneous Collaborative Perception

Congzhang Shao, Guiyang Luo, Quan Yuan, Yifu Chen, Yilin Liu, Gong Kexin, Jinglin Li

ECCV 2024 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Collaborative multi-agent perception · 14 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 node position encoding (NPE) to graph transformer (ablation)Hetecooper · OPV2V PointPillar-SECOND (hetero)AP@0.7: 0.668 -> 0.692Improved
add multi-scale fusion (MS) (ablation)Hetecooper with NPE · OPV2V PointPillar-SECOND (hetero)AP@0.7: 0.692 -> 0.725Improved
add edge-weight-guided attention (EGA) (ablation)Hetecooper with NPE and MS · OPV2V PointPillar-SECOND (hetero)AP@0.7: 0.725 -> 0.735Improved
add NPE, MS and EGA together under homogeneous models (ablation)Hetecooper · OPV2V PointPillar homogeneousAP@0.5: 0.774 -> 0.852Improved
build feature collaboration graph with feature points as nodes instead of resizing featuresheterogeneous collaborative perception · OPV2V PointPillar-VoxelNetAP@0.50: 0.788 -> 0.822 (vs HM-ViT)Improved
use Hetecooper in homogeneous settingcollaborative perception · OPV2V SECOND homogeneousAP@0.70: AP@0.50 best 0.863; AP@0.70 0.759 vs 0.783 (MDPA)Mixed
adapt fusion model trained with p2 directly to neb parameters p3 and p5Hetecooper vs MDPA · OPV2V PointPillar (hetero)AP@0.5: p5: 0.488 vs MDPA 0.219Improved
increase number of agents (ablation)Hetecooper · OPV2Vdetection accuracyMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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