AI Paper Atlas

Research map / Point clouds / Point cloud classification

AGConv: Adaptive Graph Convolution on 3D Point Clouds

Mingqiang Wei, Zeyong Wei, Haoran Zhou, Fei Hu, Huajian Si, Zhilei Chen, Zhe Zhu, Jingbo Qiu, Xuefeng Yan, Yanwen Guo, Jun Wang, Jing Qin

TPAMI 2023 · Point clouds › Point cloud classification · approach: Local feature aggregation networks · 102 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
use adaptive kernels (AGConv) instead of standard graph convolution (ablation)AGConv network · ShapeNetPartmIoU (%): 85.5 -> 86.4Improved
use AGConv instead of point-wise attention (ablation)AGConv network · ShapeNetPartmIoU (%): 83.3 -> 86.4Improved
use AGConv instead of channel-wise attention (ablation)AGConv network · ShapeNetPartmIoU (%): 83.0 -> 86.4Improved
use AGConv instead of standard graph convolution (ablation)AGConv network · ModelNet40OA (%): 92.5 -> 93.4Improved
use AGConv instead of point attention (ablation)AGConv network · ModelNet40OA (%): 92.1 -> 93.4Improved
use AGConv instead of channel attention (ablation)AGConv network · ModelNet40OA (%): 92.2 -> 93.4Improved
convolve kernel with xyz instead of features (ablation)AGConv network · ShapeNetPartmIoU (%): 85.9 -> 86.4Improved
use xyz only instead of normals only (ablation)AGConv network · ShapeNetPartmIoU (%): 86.2 -> 86.4Improved
use xyz only instead of xyz plus normals (ablation)AGConv network · ShapeNetPartmIoU (%): 86.1 -> 86.4No clear effect
add spatial transformer network (ablation)AGConv segmentation network · ShapeNetPartmIoU (%): 86.2 -> 86.4Improved
increase neighbors from k=5 to k=20 (ablation)AGConv classification network · ModelNet40OA (%): 92.8 -> 93.4Improved
increase neighbors from k=20 to k=40 (ablation)AGConv classification network · ModelNet40OA (%): 93.4 -> 93.0Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 4 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