point cloud processing

This work addresses the limited representational capacity of point cloud learning by proposing a cross-modal representation learning framework that incorporates 2D visual priors. Methodologically, it innovatively leverages pre-trained 2D vision models to guide 3D point cloud network training via feature alignment—enabling effective 2D→3D knowledge transfer without naïve modality conversion. The framework integrates a point cloud encoder-decoder architecture, self-supervised pre-training, and primitive-level supervised segmentation into a unified learning paradigm. Extensive experiments on ScanNet and S3DIS benchmarks demonstrate substantial improvements in point cloud segmentation and scene understanding performance, validating the efficacy of 2D semantic priors in enhancing 3D representation learning. The proposed approach establishes a scalable, multimodal representation learning pathway for efficient 3D perception in autonomous driving and robotics applications.

Existing quad-mesh extraction methods rely on idealized mesh-preserving parameterizations, yet real-world inputs often exhibit unmodeled geometric and topological deviations, severely compromising robustness. This work systematically characterizes typical deviation patterns from mesh preservation in non-ideal parameterizations for the first time. We propose a discrete topological operation sequence modeling framework that explicitly translates continuous mapping distortions into computable, verifiable discrete operations—including vertex splitting, edge flipping, and face re-partitioning. Our framework unifies the description of deviation causes and corresponding correction pathways, providing a principled theoretical foundation for algorithm design. Experiments demonstrate that the proposed method significantly improves fault tolerance against noise, geometric distortion, and topological inconsistencies. It achieves more stable and higher-quality quad-mesh extraction across multiple benchmark datasets.

This paper addresses the high computational complexity of two-view triangulation in multiview geometry. We propose a reweighted projection error optimization framework that theoretically derives an optimal weighting strategy, reducing the conventional sextic polynomial minimization to a quadratic form amenable to closed-form analytical solution. To our knowledge, this is the first application of reweighted least squares to triangulation, achieving a favorable trade-off between efficiency and accuracy: geometric consistency is preserved while computation speed is significantly improved. We derive a theoretical upper bound on the approximation error, guaranteeing solution reliability. Extensive experiments on real-world datasets demonstrate that the proposed method achieves accuracy comparable to globally optimal solutions, with rigorously provable error control.

This paper presents a systematic review of learning-based 3D representations for tasks such as 3D reconstruction, novel view synthesis, and rendering, spanning from traditional explicit formulations—including meshes, point clouds, and voxels—to emerging implicit neural fields and primitive-based splatting techniques like 3D Gaussian Splatting. Emphasizing the evolutionary trajectory of 3D representations themselves, the work highlights the paradigm shift from explicit to implicit modeling, clarifying the mathematical formulations, strengths, limitations, and suitable application scenarios of each approach. Distinct from prior task-centric surveys, this study centers on representation as the core organizing principle, offering fresh insights for 3D/4D content generation and identifying key challenges and future research directions to serve as a theoretical reference for the computer graphics and vision communities.

Existing unsupervised point cloud recognition methods are constrained by unimodal feature modeling, struggling to jointly capture geometric structure and semantic meaning, and exhibit poor generalization under few-shot settings. To address this, we propose the first training-free dual-branch point cloud recognition framework that synergistically models geometry and semantics via non-parametric geometric feature extraction and CLIP-style text-point-cloud semantic alignment. We introduce two novel components: (i) a Geometric Feature Enhancement (GFE) module that strengthens structural representation through learnable geometric priors, and (ii) a Multi-domain Few-shot Feature Calibration (MFF) module that improves cross-domain generalization by adaptively aligning sparse support features. Extensive experiments on ModelNet-40 and ScanObjectNN demonstrate that our method consistently outperforms all existing training-free approaches, achieving new state-of-the-art performance in both zero-shot and few-shot recognition benchmarks.

This work addresses the limitations of conventional 3D printing pipelines that rely on watertight meshes, which are prone to geometric or topological errors when reconstructed from point clouds, often leading to print failures. To overcome this, the authors propose PrintAnything, the first end-to-end framework capable of directly generating printable G-code from raw point clouds without requiring mesh reconstruction. The method projects 3D points onto slicing planes to produce a 2D representation aligned with layer-by-layer fabrication and introduces a novel G-plan graph encoding occupancy, regional segmentation, and flow-field information to jointly represent geometry and extrusion instructions. This unified representation enables a deep learning model to synthesize toolpaths directly. Experiments demonstrate that the approach accurately produces functional G-code from unprocessed point clouds, effectively bypassing reconstruction-induced defects and establishing a robust, mesh-free 3D printing pipeline.

The field of 3D vision suffers from fragmented data representations, learning paradigms, and benchmarking protocols, leading to a lack of unified understanding regarding efficiency, fidelity, and scalability. This work proposes the first cohesive conceptual framework that integrates geometric representations—such as point clouds, meshes, voxels, and 3D Gaussians—with diverse learning paradigms—including 2D-supervised learning, implicit neural representations, and 4D modeling—and connects them to real-world application scenarios. By constructing a structured knowledge graph of 3D vision, the study systematically relates dataset design, supervision mechanisms, and task requirements, clarifying the trade-offs between efficiency and fidelity and charting pathways for multimodal geometric grounding. This framework offers systematic guidance for reconstruction, generation, and dynamic scene modeling, advancing the field toward a unified and efficient paradigm.

This work addresses the challenge in 3D dataset pruning where the long-tailed class distribution induces a conflict between optimizing overall accuracy (OA) and mean per-class accuracy (mAcc), making it difficult for existing methods to simultaneously preserve representativeness and align with evaluation metrics. The authors formulate pruning as approximating the expected risk of the full dataset via a weighted subset, introducing a class-wise retention quota mechanism to ensure adequate coverage of tail classes. By integrating representation-aware subset selection with prior-invariant teacher supervision, the method enables soft-label calibration and embedding geometry distillation. This approach is the first to systematically mitigate coverage error and prior mismatch bias in 3D pruning, achieving significant improvements in both OA and mAcc across multiple 3D datasets while flexibly accommodating downstream task preferences, thereby demonstrating its effectiveness and generalizability.

Existing CAD learning approaches discretize B-Rep models into triangle meshes, thereby discarding the analytical surface representations and topological information essential for consistent instance-level analysis. This work proposes STEP-Parts, a deterministic pipeline that directly extracts geometric instance partitions from native STEP B-Rep data. The method defines partitions based on intrinsic B-Rep topology, merges faces using analytical surface types and near-tangent plane continuity criteria, and transfers labels to triangulated meshes via face-to-mesh correspondence mapping. STEP-Parts ensures boundary consistency across varying triangulations and processes the DeepCAD subset of the ABC dataset—comprising approximately 180,000 models—in under six hours. The resulting labels significantly enhance performance in implicit reconstruction-segmentation tasks and point cloud networks. Code and precomputed labels are publicly released.

Conventional geometric accuracy monitoring of unstructured point cloud data (PCD) suffers from error accumulation, high computational cost, and artifact generation due to reliance on registration and mesh reconstruction. Method: This paper proposes a preprocessing-free, intrinsic geometry-aware monitoring method that directly extracts intrinsic geometric features from raw PCD via Laplacian eigenmaps and geodesic distance computation. It incorporates an unsupervised threshold selection mechanism and a dual-strategy feature learning framework to enable robust defect identification. Contribution/Results: Experimental results demonstrate that the method achieves efficient and accurate detection of diverse geometric defects—including dents, bulges, and misalignments—without registration or reconstruction. It maintains high precision (>96% F1-score) and strong stability across heterogeneous manufacturing scenarios (e.g., additive manufacturing, CNC machining), reducing monitoring time by up to 78% compared to state-of-the-art approaches while eliminating reconstruction-induced artifacts. The framework significantly enhances both efficiency and reliability in industrial PCD-based quality inspection.