PatchAlign3D: Local Feature Alignment for Dense 3D Shape understanding
- URL: http://arxiv.org/abs/2601.02457v1
- Date: Mon, 05 Jan 2026 18:55:45 GMT
- Title: PatchAlign3D: Local Feature Alignment for Dense 3D Shape understanding
- Authors: Souhail Hadgi, Bingchen Gong, Ramana Sundararaman, Emery Pierson, Lei Li, Peter Wonka, Maks Ovsjanikov,
- Abstract summary: Current foundation models for 3D shapes excel at global tasks (retrieval, classification) but transfer poorly to local part-level reasoning.<n>We introduce an encoder-only 3D model that produces language-aligned patch-level features directly from point clouds.<n>Our 3D encoder achieves zero-shot 3D part segmentation with fast single-pass inference without any test-time multi-view rendering.
- Score: 67.15800065888887
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Current foundation models for 3D shapes excel at global tasks (retrieval, classification) but transfer poorly to local part-level reasoning. Recent approaches leverage vision and language foundation models to directly solve dense tasks through multi-view renderings and text queries. While promising, these pipelines require expensive inference over multiple renderings, depend heavily on large language-model (LLM) prompt engineering for captions, and fail to exploit the inherent 3D geometry of shapes. We address this gap by introducing an encoder-only 3D model that produces language-aligned patch-level features directly from point clouds. Our pre-training approach builds on existing data engines that generate part-annotated 3D shapes by pairing multi-view SAM regions with VLM captioning. Using this data, we train a point cloud transformer encoder in two stages: (1) distillation of dense 2D features from visual encoders such as DINOv2 into 3D patches, and (2) alignment of these patch embeddings with part-level text embeddings through a multi-positive contrastive objective. Our 3D encoder achieves zero-shot 3D part segmentation with fast single-pass inference without any test-time multi-view rendering, while significantly outperforming previous rendering-based and feed-forward approaches across several 3D part segmentation benchmarks. Project website: https://souhail-hadgi.github.io/patchalign3dsite/
Related papers
- Evaluating Foundation Models' 3D Understanding Through Multi-View Correspondence Analysis [38.10984626023432]
We introduce a novel benchmark for in-context 3D scene understanding that requires no finetuning and directly probes the quality of dense visual features.<n>We benchmark 8 state-of-the-art foundation models and show DINO-based encoders remain competitive across large viewpoint shifts.
arXiv Detail & Related papers (2025-12-12T14:03:16Z) - TriCLIP-3D: A Unified Parameter-Efficient Framework for Tri-Modal 3D Visual Grounding based on CLIP [52.79100775328595]
3D visual grounding allows an embodied agent to understand visual information in real-world 3D environments based on human instructions.<n>Existing 3D visual grounding methods rely on separate encoders for different modalities.<n>We propose a unified 2D pre-trained multi-modal network to process all three modalities.
arXiv Detail & Related papers (2025-07-20T10:28:06Z) - Pts3D-LLM: Studying the Impact of Token Structure for 3D Scene Understanding With Large Language Models [9.658828841170472]
This work presents a rigorous study of 3D token structures, systematically comparing video-based and point-based representations.<n>We propose a novel approach that enriches visual tokens by incorporating 3D point cloud features from a Sonata pretrained Point Transformer V3 encoder.
arXiv Detail & Related papers (2025-06-06T02:35:26Z) - 3D Part Segmentation via Geometric Aggregation of 2D Visual Features [57.20161517451834]
Supervised 3D part segmentation models are tailored for a fixed set of objects and parts, limiting their transferability to open-set, real-world scenarios.<n>Recent works have explored vision-language models (VLMs) as a promising alternative, using multi-view rendering and textual prompting to identify object parts.<n>To address these limitations, we propose COPS, a COmprehensive model for Parts that blends semantics extracted from visual concepts and 3D geometry to effectively identify object parts.
arXiv Detail & Related papers (2024-12-05T15:27:58Z) - Unified Scene Representation and Reconstruction for 3D Large Language Models [40.693839066536505]
Existing approaches extract point clouds either from ground truth (GT) geometry or 3D scenes reconstructed by auxiliary models.
We introduce Uni3DR2 extracts 3D geometric and semantic aware representation features via the frozen 2D foundation models.
Our learned 3D representations not only contribute to the reconstruction process but also provide valuable knowledge for LLMs.
arXiv Detail & Related papers (2024-04-19T17:58:04Z) - DatasetNeRF: Efficient 3D-aware Data Factory with Generative Radiance Fields [68.94868475824575]
This paper introduces a novel approach capable of generating infinite, high-quality 3D-consistent 2D annotations alongside 3D point cloud segmentations.
We leverage the strong semantic prior within a 3D generative model to train a semantic decoder.
Once trained, the decoder efficiently generalizes across the latent space, enabling the generation of infinite data.
arXiv Detail & Related papers (2023-11-18T21:58:28Z) - PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained
Image-Language Models [56.324516906160234]
Generalizable 3D part segmentation is important but challenging in vision and robotics.
This paper explores an alternative way for low-shot part segmentation of 3D point clouds by leveraging a pretrained image-language model, GLIP.
We transfer the rich knowledge from 2D to 3D through GLIP-based part detection on point cloud rendering and a novel 2D-to-3D label lifting algorithm.
arXiv Detail & Related papers (2022-12-03T06:59:01Z) - MvDeCor: Multi-view Dense Correspondence Learning for Fine-grained 3D
Segmentation [91.6658845016214]
We propose to utilize self-supervised techniques in the 2D domain for fine-grained 3D shape segmentation tasks.
We render a 3D shape from multiple views, and set up a dense correspondence learning task within the contrastive learning framework.
As a result, the learned 2D representations are view-invariant and geometrically consistent.
arXiv Detail & Related papers (2022-08-18T00:48:15Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.