SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1801.02108 · 2018

SBNet: Sparse Blocks Network for Fast Inference

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

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 without noticeable loss of accuracy.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("1801.02108")
get_code_for_paper("1801.02108")
have("1801.02108")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/1801.02108.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/1801.02108.svg)](https://syntology.ai/paper/1801.02108)

Connect an agent — have() is free.