Embracing Single Stride 3D Object Detector with Sparse Transformer
This addresses a key bottleneck in autonomous driving by improving detection of small objects like pedestrians, though it is an incremental advance over existing transformer-based methods.
The paper tackles the problem of information loss in LiDAR-based 3D object detection caused by downsampling in multi-stride architectures, proposing a single-stride sparse transformer that achieves state-of-the-art results, including 83.8 LEVEL 1 AP for pedestrian detection on the Waymo Open Dataset.
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. Overlooking 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. Codes will be released at https://github.com/TuSimple/SST