AI Paper Atlas

Research map / Point clouds / Point cloud classification

CLAP: Unsupervised 3D Representation Learning for Fusion 3D Perception via Curvature Sampling and Prototype Learning

Runjian Chen, Hang Zhang, Avinash Ravichandran, Hyoungseob Park, Wenqi Shao, Alex Wong, Ping Luo

ICLR 2026 · Point clouds › Point cloud classification · approach: Point-language and open-vocabulary 3D · 1 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 joint pre-training with uniform range-guided sampling (ablation)UniPAD · NuScenes (5% data)mAP: 49.81 -> 49.55No clear effect
add Curvature Sampling to joint pre-training (ablation)CLAP · NuScenes (5% data)mAP: 49.55 -> 50.81Improved
add Prototype Learning (ablation)CLAP · NuScenes (5% data)mAP: 50.81 -> 51.17Improved
pre-train with CLAP instead of random initBEVFusion · NuScenes (5% data)mAP: 48.69 -> 51.17Improved
pre-train with ALSO instead of random initBEVFusion · NuScenes (5% data)mAP: 48.69 -> 45.34Hurt
pre-train with CLAP, fine-tune on 0.5% dataBEVFusion · NuScenes (0.5% data)mAP: 16.49 -> 23.71Improved
pre-train with CLAP, fine-tune on 2.5% dataBEVFusion · NuScenes (2.5% data)mAP: 39.12 -> 42.86Improved
pre-train with CLAP, fine-tune on 1% dataBEVFusion · NuScenes (1% data)mAP: 26.22 -> 30.53Improved
pre-train with CLAP instead of random initCenterPoint · Waymo (Level-1)mAP: 61.60 -> 62.87Improved
pre-train with UniPAD instead of random initCenterPoint · Waymo (Level-1)mAP: 61.60 -> 61.57No clear effect
pre-train with CLAP instead of UniPADCylinder3D · SemanticKITTImIoU: 31.88 -> 34.28Improved
pre-train on NuScenes with CLAP, transfer to ONCECenterPoint · ONCE (5% data)mAP: 46.07 -> 46.88Improved

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