AI Paper Atlas

Research map / Detection and segmentation / 3D detection and driving

Graph R-CNN: Towards Accurate 3D Object Detection with Semantic-Decorated Local Graph

Honghui Yang, Zili Liu, Xiaopei Wu, Wenxiao Wang, Wei Qian, Xiaofei He, Deng Cai

ECCV 2022 · Detection and segmentation › 3D detection and driving · approach: Unified LiDAR 3D detection · 104 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 RoI-graph pooling (RGP) (ablation)CenterPoint-Voxel baseline · WOD val vehicle LEVEL 23D APH: 66.17 -> 69.48Improved
add RoI-graph pooling (RGP) (ablation)CenterPoint-Voxel baseline · WOD val pedestrian LEVEL 23D APH: 62.67 -> 66.79Improved
add multi-level attentive fusion (MLAF) (ablation)Graph R-CNN with RGP · WOD val vehicle LEVEL 23D APH: 69.48 -> 69.77Improved
add dynamic point aggregation (DPA) (ablation)Graph R-CNN with RGP and MLAF · WOD val vehicle LEVEL 23D APH: 69.77 -> 70.83Improved
add dynamic point aggregation (DPA) (ablation)Graph R-CNN with RGP and MLAF · WOD val vehicle LEVEL 2 (0-30m)3D APH: 90.25 -> 91.74Improved
add PointNet encoding of neighbor points (ablation)Graph R-CNN · WOD val vehicle LEVEL 23D APH: 70.83 -> 71.04Improved
add two diagonal corners per graph node (ablation)Graph R-CNN · WOD val vehicle LEVEL 23D APH: 71.04 -> 72.10Improved
use patch search instead of point cloud region pooling for grouping (ablation)Graph R-CNN · WOD valruntime (ms): 69.7 -> 1.9Improved
increase graph iterations from T=1 to T=3 (ablation)Graph R-CNN · WOD val vehicle LEVEL 23D APH: 68.76 -> 69.48Improved
replace channel-wise Transformer aggregation with max pooling (ablation)Graph R-CNN · WOD val vehicle LEVEL 23D APH: 71.67 -> 72.10Improved
use DFVS instead of FPS for sampling (ablation)3DSSD · KITTI val car3D mAP / runtime: 81.73 -> 81.75; 29.6 -> 20.3No clear effect
add image features (2D detector output) to local graphs (ablation)Graph R-CNN · KITTI val car3D AP: 85.38 -> 86.87Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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