Conventional deep convolutional neural networks (CNNs) apply convolution operators uniformly in space across all feature maps for hundreds of layers - this incurs a high computational cost for real time applications. For many problems such as object detection and semantic segmentation, we are able to obtain a low-cost computation mask, either from a priori problem knowledge, or from a low resolution segmentation network. We show that such computation masks can be used to reduce computation in the high resolution main network. Variants of sparse activation CNNs have previously been explored on small scale tasks, and showed no degradation in terms of object classification accuracy, but often measured gains in terms of theoretical FLOPs without realizing a practical speed-up when compared to highly optimized dense convolution implementations. In this work, we leverage the sparsity structure of computation masks and propose a novel tiling-based sparse convolution algorithm. We verified the effectiveness of our sparse CNN on LiDAR based 3D object detection, and we report significant wall-clock speed-ups compared to dense convolution, as well as improved detection accuracy.

本文介绍了一种新型的基于拼贴的稀疏卷积算法，通过利用计算掩码的稀疏结构，降低了CNN中高分辨率计算的复杂度，并应用于基于LiDAR的3D目标检测中，最终获得了显著的速度提升而无需牺牲准确率。

SBNet: 用于快速推断的稀疏块网络