DA-Occ: Efficient 3D Voxel Occupancy Prediction via Directional 2D for Geometric Structure Preservation

2025-07-31

AI Summary

To address the dual requirements of accuracy and inference speed for 3D voxel occupancy prediction in autonomous driving, this paper proposes a lightweight and efficient directional 2D feature modeling framework. The method preserves vertical geometric structure via directional feature slicing, recovers height cues by fusing multi-view 2D features using a novel directional attention mechanism, and incorporates geometric-aware design in the bird’s-eye view (BEV) space—eliminating the computational overhead of explicit 3D convolutions. Evaluated on Occ3D-nuScenes, our approach achieves 39.3% mIoU at 27.7 FPS on GPU and 14.8 FPS on edge devices, substantially outperforming existing real-time methods. It establishes a new Pareto-optimal trade-off between accuracy and efficiency while maintaining 3D structural integrity.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Efficient and high-accuracy 3D occupancy prediction is crucial for ensuring the performance of autonomous driving (AD) systems. However, many current methods focus on high accuracy at the expense of real-time processing needs. To address this challenge of balancing accuracy and inference speed, we propose a directional pure 2D approach. Our method involves slicing 3D voxel features to preserve complete vertical geometric information. This strategy compensates for the loss of height cues in Bird's-Eye View (BEV) representations, thereby maintaining the integrity of the 3D geometric structure. By employing a directional attention mechanism, we efficiently extract geometric features from different orientations, striking a balance between accuracy and computational efficiency. Experimental results highlight the significant advantages of our approach for autonomous driving. On the Occ3D-nuScenes, the proposed method achieves an mIoU of 39.3% and an inference speed of 27.7 FPS, effectively balancing accuracy and efficiency. In simulations on edge devices, the inference speed reaches 14.8 FPS, further demonstrating the method's applicability for real-time deployment in resource-constrained environments.

Problem

• Balancing 3D occupancy prediction accuracy and real-time processing
• Preserving vertical geometry in 2D slices for 3D structure integrity
• Optimizing directional attention for efficient feature extraction

Innovation

• Directional 2D approach for 3D occupancy prediction
• Slicing 3D voxel features preserves vertical geometry
• Directional attention balances accuracy and efficiency