EA-LSS: Edge-aware Lift-splat-shot Framework for 3D BEV Object Detection
It improves accuracy for camera-only and multi-model 3D object detection in autonomous driving, but is incremental as it builds on existing LSS methods.
The paper tackles the 'depth jump' problem in LSS-based 3D object detection by proposing an edge-aware framework, achieving state-of-the-art performance with 76.5% mAP and 77.6% NDS on nuScenes benchmarks.
In recent years, great progress has been made in the Lift-Splat-Shot-based (LSS-based) 3D object detection method. However, inaccurate depth estimation remains an important constraint to the accuracy of camera-only and multi-model 3D object detection models, especially in regions where the depth changes significantly (i.e., the "depth jump" problem). In this paper, we proposed a novel Edge-aware Lift-splat-shot (EA-LSS) framework. Specifically, edge-aware depth fusion (EADF) module is proposed to alleviate the "depth jump" problem and fine-grained depth (FGD) module to further enforce refined supervision on depth. Our EA-LSS framework is compatible for any LSS-based 3D object detection models, and effectively boosts their performances with negligible increment of inference time. Experiments on nuScenes benchmarks demonstrate that EA-LSS is effective in either camera-only or multi-model models. It is worth mentioning that EA-LSS achieved the state-of-the-art performance on nuScenes test benchmarks with mAP and NDS of 76.5% and 77.6%, respectively.