In this paper, we propose a long-sequence modeling framework, named StreamPETR, for multi-view 3D object detection. Built upon the sparse query design in the PETR series, we systematically develop an object-centric temporal mechanism. The model is performed in an online manner and the long-term historical information is propagated through object queries frame by frame. Besides, we introduce a motion-aware layer normalization to model the movement of the objects. StreamPETR achieves significant performance improvements only with negligible computation cost, compared to the single-frame baseline. On the standard nuScenes benchmark, it reaches a new state-of-the-art performance (63.6% NDS). The lightweight version realizes 45.0% mAP and 31.7 FPS, outperforming the state-of-the-art method (SOLOFusion) by 2.3% mAP and 1.8x faster FPS. Code will be available at https://github.com/exiawsh/StreamPETR.git.

翻译：本文提出一种名为StreamPETR的长序列建模框架，用于多视角3D目标检测。该框架基于PETR系列的稀疏查询设计，系统性地开发了以物体为中心的时序机制。模型以在线方式运行，通过逐帧传递物体查询来传播长期历史信息。此外，我们引入运动感知层归一化以建模物体运动。与单帧基线相比，StreamPETR在仅增加微不足道计算成本的情况下实现了显著的性能提升。在标准nuScenes基准测试中，该方法达到了最新的最优性能（63.6% NDS）。轻量级版本实现了45.0% mAP和31.7 FPS，较当前最优方法SOLOFusion分别提升了2.3% mAP和1.8倍帧率。代码将发布于https://github.com/exiawsh/StreamPETR.git。