AI Paper Atlas

Research map / Point clouds / Point cloud classification

DRINet: A Dual-Representation Iterative Learning Network for Point Cloud Segmentation

Maosheng Ye, Shuangjie Xu, Tongyi Cao, Qifeng Chen

ICCV 2021 · Point clouds › Point cloud classification · approach: Local feature aggregation networks · 50 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 SVPFE (voxel branch) to a point-only network (ablation)DRINet · SemanticKITTI valmIoU: 48.6 -> 67.3Improved
add SPVFE (point branch) to a voxel-only network (ablation)DRINet · SemanticKITTI valmIoU: 63.4 -> 67.3Improved
add multi-scale pooling layer (ablation)DRINet · SemanticKITTI valmIoU: 65.4 -> 67.3Improved
add attentive gathering instead of nearest gathering (ablation)DRINet · SemanticKITTI valmIoU: 64.6 -> 65.4Improved
increase SPVFE and SVPFE blocks from 3 to 4 (ablation)DRINet · SemanticKITTI valmIoU: 67.3 -> 67.6, inference 62ms -> 72msMixed
increase block number from 1 to 4 (ablation)DRINet · SemanticKITTI valmIoU: 58.2% to 67.6%Improved
increase block number from 1 to 4 (ablation)DRINet · S3DISmIoU: 36.3% to 66.7%Improved
use GAFE geometry-aware features instead of PointNet++ features (ablation)feature extractor · SemanticKITTI valmIoU: 18.1 -> 22.8Improved
use DRINet instead of SPVNetSemanticKITTI segmentation · SemanticKITTI testmIoU / speed (ms): 67.0 -> 67.5 mIoU, 259 -> 62 msImproved
use DRINet instead of PVCNNShapeNet Parts segmentation · ShapeNet PartsmIoU / latency (ms): 86.2 -> 86.4 mIoU, 50.7 -> 30 msImproved
use DRINet instead of KPConvS3DIS Area 5 · S3DIS Area 5mIoU: 67.1 -> 66.7Hurt

Later papers that compared with this method

1 later accepted paper reports DRINet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

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