FSHNet: Fully Sparse Hybrid Network for 3D Object Detection
This work addresses efficiency and accuracy issues in 3D object detection for autonomous driving systems, representing an incremental improvement over existing sparse detectors.
The paper tackles the problem of fully sparse 3D object detectors, which suffer from impaired long-range interactions and center feature missing, by introducing FSHNet with a SlotFormer block, dynamic sparse label assignment, and sparse upsampling, achieving state-of-the-art results on Waymo, nuScenes, and Argoverse2 benchmarks.
Fully sparse 3D detectors have recently gained significant attention due to their efficiency in long-range detection. However, sparse 3D detectors extract features only from non-empty voxels, which impairs long-range interactions and causes the center feature missing. The former weakens the feature extraction capability, while the latter hinders network optimization. To address these challenges, we introduce the Fully Sparse Hybrid Network (FSHNet). FSHNet incorporates a proposed SlotFormer block to enhance the long-range feature extraction capability of existing sparse encoders. The SlotFormer divides sparse voxels using a slot partition approach, which, compared to traditional window partition, provides a larger receptive field. Additionally, we propose a dynamic sparse label assignment strategy to deeply optimize the network by providing more high-quality positive samples. To further enhance performance, we introduce a sparse upsampling module to refine downsampled voxels, preserving fine-grained details crucial for detecting small objects. Extensive experiments on the Waymo, nuScenes, and Argoverse2 benchmarks demonstrate the effectiveness of FSHNet. The code is available at https://github.com/Say2L/FSHNet.