CLIP2Point: Transfer CLIP to Point Cloud Classification with Image-Depth Pre-training
This work addresses the challenge of limited pre-training data for 3D vision, enabling better point cloud classification for applications in robotics and autonomous systems, though it is incremental by building on existing CLIP-based approaches.
The paper tackles the problem of transferring vision-language pre-training models like CLIP to 3D point cloud classification by addressing domain gaps between rendered depth maps and images, proposing CLIP2Point with image-depth pre-training and a Dual-Path Adapter module. It achieves state-of-the-art results in zero-shot and few-shot classification, outperforming PointCLIP and other methods.
Pre-training across 3D vision and language remains under development because of limited training data. Recent works attempt to transfer vision-language pre-training models to 3D vision. PointCLIP converts point cloud data to multi-view depth maps, adopting CLIP for shape classification. However, its performance is restricted by the domain gap between rendered depth maps and images, as well as the diversity of depth distributions. To address this issue, we propose CLIP2Point, an image-depth pre-training method by contrastive learning to transfer CLIP to the 3D domain, and adapt it to point cloud classification. We introduce a new depth rendering setting that forms a better visual effect, and then render 52,460 pairs of images and depth maps from ShapeNet for pre-training. The pre-training scheme of CLIP2Point combines cross-modality learning to enforce the depth features for capturing expressive visual and textual features and intra-modality learning to enhance the invariance of depth aggregation. Additionally, we propose a novel Dual-Path Adapter (DPA) module, i.e., a dual-path structure with simplified adapters for few-shot learning. The dual-path structure allows the joint use of CLIP and CLIP2Point, and the simplified adapter can well fit few-shot tasks without post-search. Experimental results show that CLIP2Point is effective in transferring CLIP knowledge to 3D vision. Our CLIP2Point outperforms PointCLIP and other self-supervised 3D networks, achieving state-of-the-art results on zero-shot and few-shot classification.