AI Paper Atlas

Research map / Point clouds / Segmentation and detection

Implicit Surface Contrastive Clustering for LiDAR Point Clouds

Zaiwei Zhang, Min Bai, Erran Li

CVPR 2023 · Point clouds › Segmentation and detection · approach: LiDAR segmentation for driving · 7 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
pretrain with ISCC instead of SSPL3D sparse U-Net · SemanticKITTI, 1% fine-tuningmIoU: 42.5 -> 45.1Improved
pretrain with ISCC instead of SSPL3D sparse U-Net · Waymo Open Dataset, 1% fine-tuningmIoU: 44.8 -> 46.0Improved
pretrain with ISCC instead of DepthContrastPart-A2 3D detector, KITTI Car Moderate · KITTI, 5% fine-tuningmAP R11: 65.0 -> 68.9Improved
add occupancy prediction with 8 neighboring voxel features (ablation)occupancy-only pretraining · SemanticKITTI, 1% fine-tuningmIoU: 38.9 -> 40.5Improved
increase occupancy aggregation from 8 to 16 neighboring voxel features (ablation)occupancy-only pretraining · SemanticKITTI, 1% fine-tuningmIoU: 40.5 -> 43.0Improved
increase occupancy aggregation from 16 to 32 neighboring voxel features (ablation)occupancy-only pretraining · SemanticKITTI, 1% fine-tuningmIoU: 43.0 -> 43.6Improved
add multi-view frames with 2-frame separation instead of single-view input (ablation)ISCC · SemanticKITTI, 1% fine-tuningmIoU: 43.7 -> 44.0Improved
increase multi-view frame separation from 2 to 5 (ablation)ISCC · SemanticKITTI, 1% fine-tuningmIoU: 44.0 -> 45.1Improved
add the global clustering loss to local-only contrastive learning (ablation)ISCC · SemanticKITTI, 1% fine-tuningmIoU: 41.3 -> 43.6Improved
add the local contrastive loss to global-only clustering (ablation)ISCC · SemanticKITTI, 1% fine-tuningmIoU: 41.7 -> 43.6Improved
increase the maximum cluster count from 100 to 200 (ablation)full contrastive clustering · SemanticKITTI, 1% fine-tuningmIoU: 43.6 -> 43.5No clear effect

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