Embracing Single Stride 3D Object Detector with Sparse Transformer

By Lue Fan, Ziqi Pang, Tianyuan Zhang, Yu-Xiong Wang, Hang Zhao, Feng Wang, Naiyan Wang, Zhaoxiang Zhang. Computer Vision and Pattern Recognition, 2021

351 citations recorded by the source corpus.

In LiDAR-based 3D object detection for autonomous driving, the ratio of the object size to input scene size is significantly smaller compared to 2D detection cases. Over-looking this difference, many 3D detectors directly follow the common practice of 2D detectors, which downsample the feature maps even after quantizing the point clouds. In this paper, we start by rethinking how such multi-stride stereotype affects the LiDAR-based 3D object detectors. Our experiments point out that the downsampling operations bring few advantages, and lead to inevitable information loss. To remedy this issue, we propose Single-stride Sparse Transformer (SST) to maintain the original resolution from the beginning to the end of the network. Armed with transformers, our method addresses the problem of insufficient receptive field in single-stride architectures. It also cooperates well with the sparsity of point clouds and naturally avoids expensive computation. Eventually, our SST achieves state-of-the-art results on the large-scale Waymo Open Dataset. It is worth mentioning that our method can achieve exciting performance (83.8 LEVEL_1 AP on validation split) on small object (pedestrian) detection due to the characteristic of single stride. Our codes will be public soon.

View Paper: https://www.semanticscholar.org/paper/babf62e12510a025dad4a602bd2e9f3a8331314d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A Survey of Visual Transformers (2021, background): “Fine-Grained: Pointformer [110], SST [111],”
3D Object Detection for Autonomous Driving: A Comprehensive Survey (2022, methodology): “The development of backbone networks has experienced four stages: (1) 2D networks to process BEV images that are generated by point cloud projection [10, 335], (2) 2D networks to process pillars that are generated by PointNet encoding [124], (3) 3D sparse convolutional networks to process voxelized ”
V2V4Real: A Real-World Large-Scale Dataset for Vehicle-to-Vehicle Cooperative Perception (2023, background): “Benefiting from its fast processing and real-time performance, many recent 3D object detection models follow this pillar-based approach [12, 41].”
Point Transformer V3: Simpler, Faster, Stronger (2023, background): “Our PTv3, engaged with CenterPoint, consistently outperforms both sparse convolutional [26, 102] and transformer-based [22, 28] detectors, achieving significant gains even when compared with the recent state-of-the-art, FlatFormer [51].”