Point-Focused Attention Meets Context-Scan State Space: Robust Biological Visual Perception for Point Cloud Representation

2026-10-08

AI Summary

This study addresses the challenge of jointly modeling local geometry and global context in point cloud representation learning. To this end, we propose PointLearner, a novel bio-inspired "focus-context" architecture that mimics the foveal vision and saccadic mechanisms of biological visual systems. Methodologically, PointLearner achieves fine-grained local feature extraction through point-focused attention and learnable induced point pooling, while efficiently capturing long-range dependencies by integrating competitive normalization attention, Hilbert curve-guided scanning, and a bidirectional S6 state space model. Extensive experiments demonstrate that the proposed method exhibits exceptional robustness across multiple point cloud tasks, achieving state-of-the-art performance.

Abstract

Synergistically capturing intricate local structures and global contextual dependencies has become a critical challenge in point cloud representation learning. To address this, we introduce PointLearner, a point cloud representation learning network that closely aligns with biological vision which employs an active, foveation-inspired processing strategy, thus enabling local geometric modeling and long-range dependency interactions simultaneously. Specifically, we first design a point-focused attention, which simulates foveal vision at the visual focus through a competitive normalized attention mechanism between local neighbors and spatially downsampled features. The spatially downsampled features are extracted by a pooling method based on learnable inducing points, which can flexibly adapt to the non-uniform distribution of point clouds as the number of inducing points is controlled and they interact directly with point clouds. Second, we propose a context-scan state space that mimics eye's saccade inference, which infers the overall semantic structure and spatial content in the scene through a scan path guided by the Hilbert curve for the bidirectional S6. With this focus-then-context biomimetic design, PointLearner demonstrates remarkable robustness and achieves state-of-the-art performance across multiple point cloud tasks.

Problem

• Point Cloud Representation
• Local Structure
• Global Contextual Dependencies
• Representation Learning

Innovation

• Point Cloud Representation
• Foveation-inspired Attention
• State Space Model
• Biomimetic Vision
• Hilbert Curve