AdaOcc: Adaptive-Resolution Occupancy Prediction

TL;DR

AdaOcc introduces an adaptive-resolution 3D occupancy prediction framework that balances detail and efficiency by focusing high-resolution reconstruction on regions of interest, significantly improving accuracy in urban autonomous driving scenarios.

Contribution

The paper presents AdaOcc, a novel adaptive-resolution approach combining object-centric reconstruction with holistic occupancy prediction for efficient, detailed 3D scene understanding.

Findings

Over 13% IOU improvement in close-range scenarios

More than 40% reduction in Hausdorff distance

Effective balance of resolution and computational efficiency

Abstract

Autonomous driving in complex urban scenarios requires 3D perception to be both comprehensive and precise. Traditional 3D perception methods focus on object detection, resulting in sparse representations that lack environmental detail. Recent approaches estimate 3D occupancy around vehicles for a more comprehensive scene representation. However, dense 3D occupancy prediction increases computational demands, challenging the balance between efficiency and resolution. High-resolution occupancy grids offer accuracy but demand substantial computational resources, while low-resolution grids are efficient but lack detail. To address this dilemma, we introduce AdaOcc, a novel adaptive-resolution, multi-modal prediction approach. Our method integrates object-centric 3D reconstruction and holistic occupancy prediction within a single framework, performing highly detailed and precise 3D reconstruction only in regions of interest (ROIs). These high-detailed 3D surfaces are represented in point clouds, thus their precision is not constrained by the predefined grid resolution of the occupancy map. We conducted comprehensive experiments on the nuScenes dataset, demonstrating significant improvements over existing methods. In close-range scenarios, we surpass previous baselines by over 13% in IOU, and over 40% in Hausdorff distance. In summary, AdaOcc offers a more versatile and effective framework for delivering accurate 3D semantic occupancy prediction across diverse driving scenarios.