DA-Occ: Direction-Aware 2D Convolution for Efficient and Geometry-Preserving 3D Occupancy Prediction
This work addresses the need for real-time, geometry-preserving 3D occupancy prediction in autonomous driving systems, offering an incremental improvement over existing methods.
The paper tackles the trade-off between accuracy and efficiency in 3D occupancy prediction for autonomous driving by proposing a pure 2D framework that introduces height-score projection and direction-aware convolution, achieving an mIoU of 39.3% and inference speeds up to 27.7 FPS.
Efficient and high-accuracy 3D occupancy prediction is crucial for ensuring the performance of autonomous driving (AD) systems. However, many existing methods involve trade-offs between accuracy and efficiency. Some achieve high precision but with slow inference speed, while others adopt purely bird's-eye-view (BEV)-based 2D representations to accelerate processing, inevitably sacrificing vertical cues and compromising geometric integrity. To overcome these limitations, we propose a pure 2D framework that achieves efficient 3D occupancy prediction while preserving geometric integrity. Unlike conventional Lift-Splat-Shoot (LSS) methods that rely solely on depth scores to lift 2D features into 3D space, our approach additionally introduces a height-score projection to encode vertical geometric structure. We further employ direction-aware convolution to extract geometric features along both vertical and horizontal orientations, effectively balancing accuracy and computational efficiency. 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.