This report presents our Le3DE2E_Occ solution for 4D Occupancy Forecasting in
Argoverse Challenges at CVPR 2023 Workshop on Autonomous Driving (WAD). Our
solution consists of a strong LiDAR-based Bird's Eye View (BEV) encoder with
temporal fusion and a two-stage decoder, which combines a DETR head and a UNet
decoder. The solution was tested on the Argoverse 2 sensor dataset to evaluate
the occupancy state 3 seconds in the future. Our solution achieved 18% lower L1
Error (3.57) than the baseline and got the 1 place on the 4D Occupancy
Forecasting task in Argoverse Challenges at CVPR 2023.

我们的 Le3DE2E_Occ 解决方案在 CVPR 2023 自动驾驶（WAD）研讨会的 Argoverse 挑战中呈现了 4D 占用预测。我们的解决方案包括一个强大的基于 LiDAR 的鸟瞰图编码器，具有时间融合和两阶段解码器，结合了 DETR 头部和 UNet 解码器。该解决方案在 Argoverse 2 传感器数据集上进行了测试，以评估未来 3 秒的占用状态。我们的解决方案的 L1 误差（3.57）比基线低 18％，并在 CVPR 2023 的 4D 占用预测任务中获得第一名。