Accelerating Large-Kernel Convolution Using Summed-Area Tables

06/26/2019
by   Linguang Zhang, et al.
5

Expanding the receptive field to capture large-scale context is key to obtaining good performance in dense prediction tasks, such as human pose estimation. While many state-of-the-art fully-convolutional architectures enlarge the receptive field by reducing resolution using strided convolution or pooling layers, the most straightforward strategy is adopting large filters. This, however, is costly because of the quadratic increase in the number of parameters and multiply-add operations. In this work, we explore using learnable box filters to allow for convolution with arbitrarily large kernel size, while keeping the number of parameters per filter constant. In addition, we use precomputed summed-area tables to make the computational cost of convolution independent of the filter size. We adapt and incorporate the box filter as a differentiable module in a fully-convolutional neural network, and demonstrate its competitive performance on popular benchmarks for the task of human pose estimation.

READ FULL TEXT

page 2

page 4

research
11/12/2018

OriNet: A Fully Convolutional Network for 3D Human Pose Estimation

In this paper, we propose a fully convolutional network for 3D human pos...
research
11/20/2018

Multi-scale aggregation of phase information for reducing computational cost of CNN based DOA estimation

In a recent work on direction-of-arrival (DOA) estimation of multiple sp...
research
05/28/2020

3D human pose estimation with adaptive receptive fields and dilated temporal convolutions

In this work, we demonstrate that receptive fields in 3D pose estimation...
research
12/25/2020

Inception Convolution with Efficient Dilation Search

Dilation convolution is a critical mutant of standard convolution neural...
research
11/30/2017

Spatially-Adaptive Filter Units for Deep Neural Networks

Classical deep convolutional networks increase receptive field size by e...
research
02/25/2020

Toward fast and accurate human pose estimation via soft-gated skip connections

This paper is on highly accurate and highly efficient human pose estimat...
research
06/15/2018

Learning Front-end Filter-bank Parameters using Convolutional Neural Networks for Abnormal Heart Sound Detection

Automatic heart sound abnormality detection can play a vital role in the...

Please sign up or login with your details

Forgot password? Click here to reset