Abstract:Point cloud completion aims to reconstruct the complete 3D shape from incomplete point clouds, and it is crucial for tasks such as 3D object detection and segmentation. Despite the continuous advances in point cloud analysis techniques, feature extraction methods are still confronted with apparent limitations. The sparse sampling of point clouds, used as inputs in most methods, often results in a certain loss of global structure information. Meanwhile, traditional local feature extraction methods usually struggle to capture the intricate geometric details. To overcome these drawbacks, we introduce PointCFormer, a transformer framework optimized for robust global retention and precise local detail capture in point cloud completion. This framework embraces several key advantages. First, we propose a relation-based local feature extraction method to perceive local delicate geometry characteristics. This approach establishes a fine-grained relationship metric between the target point and its k-nearest neighbors, quantifying each neighboring point's contribution to the target point's local features. Secondly, we introduce a progressive feature extractor that integrates our local feature perception method with self-attention. Starting with a denser sampling of points as input, it iteratively queries long-distance global dependencies and local neighborhood relationships. This extractor maintains enhanced global structure and refined local details, without generating substantial computational overhead. Additionally, we develop a correction module after generating point proxies in the latent space to reintroduce denser information from the input points, enhancing the representation capability of the point proxies. PointCFormer demonstrates state-of-the-art performance on several widely used benchmarks.

Multi-resolution Transformer-Based Point Cloud Completion Network for Intelligent Manufacture

Point Cloud Completion Cascade Optimization Network Based on Feature Fusion

Dual-scale Point Cloud Completion Network Based on High-Frequency Feature Fusion

Point Cloud Completion by Dynamic Transformer with Adaptive Neighbourhood Feature Fusion

Adaptive Recurrent Forward Network for Dense Point Cloud Completion

3DPCTN: Two 3D Local-Object Point-Cloud-Completion Transformer Networks Based on Self-Attention and Multi-Resolution

CRA-PCN: Point Cloud Completion with Intra- and Inter-level Cross-Resolution Transformers

AdaPoinTr: Diverse Point Cloud Completion With Adaptive Geometry-Aware Transformers

FPTNet: Full Point Transformer Network for Point Cloud Completion.

MLFT-Net: Point Cloud Completion Using MultiLevel Feature Transformer

Point Cloud Completion Via Skeleton-Detail Transformer

Gsformer: geometric-spatial transformer on point cloud completion

Transformer-based Point Cloud Generation Network

A Point Contextual Transformer Network for Point Cloud Completion

CompleteDT: Point Cloud Completion with Dense Augment Inference Transformers

Completedt: Point Cloud Completion with Information-Perception Transformers

PoinTr-PM: Diverse Point Cloud Completion with Geometry-Aware Transformers and Point Moving

HFF-Net: Hierarchical Feature Fusion Network for Point Cloud Generation with Point Transformers

Neighborhood Multi-compound Transformer for Point Cloud Registration

PointCFormer: a Relation-based Progressive Feature Extraction Network for Point Cloud Completion

Point Cloud Completion Network Applied to Vehicle Data