AdaOcc: Adaptive Forward View Transformation and Flow Modeling for 3D Occupancy and Flow Prediction

TL;DR

This paper introduces AdaOcc, a dual-stage framework for 3D occupancy and flow prediction that improves accuracy by adaptive view transformation and flow modeling, validated on nuScenes dataset.

Contribution

The paper presents a novel dual-stage approach with adaptive view transformation and flow modeling for enhanced 3D occupancy and flow prediction in autonomous driving.

Findings

Achieved significant accuracy improvements on nuScenes dataset.

Ranked second on the public leaderboard for the challenge.

Demonstrated robustness in real-world scenarios.

Abstract

In this technical report, we present our solution for the Vision-Centric 3D Occupancy and Flow Prediction track in the nuScenes Open-Occ Dataset Challenge at CVPR 2024. Our innovative approach involves a dual-stage framework that enhances 3D occupancy and flow predictions by incorporating adaptive forward view transformation and flow modeling. Initially, we independently train the occupancy model, followed by flow prediction using sequential frame integration. Our method combines regression with classification to address scale variations in different scenes, and leverages predicted flow to warp current voxel features to future frames, guided by future frame ground truth. Experimental results on the nuScenes dataset demonstrate significant improvements in accuracy and robustness, showcasing the effectiveness of our approach in real-world scenarios. Our single model based on Swin-Base ranks second on the public leaderboard, validating the potential of our method in advancing autonomous car perception systems.