Superpixel Segmentation with Fully Convolutional Networks

03/29/2020
by   Fengting Yang, et al.
6

In computer vision, superpixels have been widely used as an effective way to reduce the number of image primitives for subsequent processing. But only a few attempts have been made to incorporate them into deep neural networks. One main reason is that the standard convolution operation is defined on regular grids and becomes inefficient when applied to superpixels. Inspired by an initialization strategy commonly adopted by traditional superpixel algorithms, we present a novel method that employs a simple fully convolutional network to predict superpixels on a regular image grid. Experimental results on benchmark datasets show that our method achieves state-of-the-art superpixel segmentation performance while running at about 50fps. Based on the predicted superpixels, we further develop a downsampling/upsampling scheme for deep networks with the goal of generating high-resolution outputs for dense prediction tasks. Specifically, we modify a popular network architecture for stereo matching to simultaneously predict superpixels and disparities. We show that improved disparity estimation accuracy can be obtained on public datasets.

READ FULL TEXT

page 3

page 7

page 8

page 12

page 13

page 14

page 15

page 16

research
02/28/2017

3D Shape Segmentation via Shape Fully Convolutional Networks

We propose a novel fully convolutional network architecture for shapes, ...
research
03/15/2017

Texture segmentation with Fully Convolutional Networks

In the last decade, deep learning has contributed to advances in a wide ...
research
02/02/2017

A Fast and Compact Saliency Score Regression Network Based on Fully Convolutional Network

Visual saliency detection aims at identifying the most visually distinct...
research
03/24/2021

Vision Transformers for Dense Prediction

We introduce dense vision transformers, an architecture that leverages v...
research
12/31/2016

Improved Stereo Matching with Constant Highway Networks and Reflective Confidence Learning

We present an improved three-step pipeline for the stereo matching probl...
research
11/20/2019

Shift Convolution Network for Stereo Matching

In this paper, we present Shift Convolution Network (ShiftConvNet) to pr...
research
05/02/2021

AG-CUResNeSt: A Novel Method for Colon Polyp Segmentation

Colorectal cancer is among the most common malignancies and can develop ...

Please sign up or login with your details

Forgot password? Click here to reset