AI Paper Atlas

Research map / Point clouds / Point cloud classification

CLIP2Point: Transfer CLIP to Point Cloud Classification with Image-Depth Pre-Training

Tianyu Huang, Bowen Dong, Yunhan Yang, Xiaoshui Huang, Rynson W.H. Lau, Wanli Ouyang, Wangmeng Zuo

ICCV 2023 · Point clouds › Point cloud classification · approach: Point-language and open-vocabulary 3D · 193 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 image-depth contrastive pre-training of the depth encoder (ablation)CLIP2Point zero-shot · ModelNet10 zero-shotaccuracy (%): 30.51 -> 66.63Improved
add pre-training (ablation)CLIP2Point zero-shot · ScanObjectNN OBJ_BG and PB_T50_RS zero-shotaccuracy (%): 23.92 -> 30.46 (OBJ_BG); 18.18 -> 23.32 (PB_T50_RS)Improved
use new depth rendering (minimum depth, dilation) instead of PointCLIP rendering (ablation)CLIP2Point without pre-training · ScanObjectNN OBJ_ONLY zero-shotaccuracy (%): 15.15 -> 28.40Improved
add intra-modality contrastive learning to cross-modality pre-training (ablation)CLIP2Point pre-training · zero-shot classificationaccuracy (%): 38.29% -> 11.09% higherImproved
use dual-path with global-view aggregator instead of single path (ablation)GDPA · ModelNet40 16-shotaccuracy (%): 87.32 -> 89.21Improved
add gating weights to dual-path fusion (ablation)GDPA · ModelNet40 16-shotaccuracy (%): 89.21 -> 89.79Improved
replace inter-view adapter with global-view aggregator (single path) (ablation)GDPA · ModelNet40 16-shotaccuracy (%): 86.06 -> 87.32Improved
add dual-path to inter-view adapter (ablation)GDPA · ModelNet40 16-shotaccuracy (%): 86.06 -> 86.18No clear effect
initialize both dual paths with CLIP encoder instead of CLIP2Point depth encoder (ablation)GDPA · ModelNet40 16-shotaccuracy (%): 89.79 -> 87.20Hurt
use minimum depth value instead of weighted summation (ablation)depth rendering · ModelNet40 zero-shot (R=2)accuracy (%): 17.63 -> 29.71Improved
increase dilation rate from 2 to 4 (ablation)depth rendering with minimum depth · ModelNet40 zero-shotaccuracy (%): 29.71 -> 28.36Hurt
replace 2D depth encoder with 3D Point Transformer encoder in pre-training (ablation)CLIP2Point · ModelNet40 zero-shotaccuracy (%): 49.38 -> 20.83%Hurt

Later papers that compared with this method

2 later accepted papers report Clip2point as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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