Technical Report for Argoverse Challenges on Unified Sensor-based Detection, Tracking, and Forecasting
This work addresses the challenge of integrating multiple perception tasks for autonomous vehicles, but it appears incremental as it builds on existing BEV and multi-task approaches.
The paper tackled the problem of unified sensor-based detection, tracking, and forecasting for autonomous driving by proposing a unified network with a BEV encoder and spatial-temporal fusion, achieving first place in all three tasks on the Argoverse 2 dataset at the CVPR 2023 workshop.
This report presents our Le3DE2E solution for unified sensor-based detection, tracking, and forecasting in Argoverse Challenges at CVPR 2023 Workshop on Autonomous Driving (WAD). We propose a unified network that incorporates three tasks, including detection, tracking, and forecasting. This solution adopts a strong Bird's Eye View (BEV) encoder with spatial and temporal fusion and generates unified representations for multi-tasks. The solution was tested in the Argoverse 2 sensor dataset to evaluate the detection, tracking, and forecasting of 26 object categories. We achieved 1st place in Detection, Tracking, and Forecasting on the E2E Forecasting track in Argoverse Challenges at CVPR 2023 WAD.