Efficient Convolutional Neural Network with Binary Quantization Layer

TL;DR

This paper presents a real-time, semantic image segmentation method using a novel binary quantization layer in CNNs, outperforming previous non-semantic segmentation approaches.

Contribution

Introduces a binary encoding layer in CNNs for semantic segmentation, enabling real-time processing and broad category applicability.

Findings

Outperforms state-of-the-art non-semantic segmentation methods

Enables real-time semantic segmentation

First to apply CNN-based semantic segmentation beyond limited categories

Abstract

In this paper we introduce a novel method for segmentation that can benefit from general semantics of Convolutional Neural Network (CNN). Our segmentation proposes visually and semantically coherent image segments. We use binary encoding of CNN features to overcome the difficulty of the clustering on the high-dimensional CNN feature space. These binary encoding can be embedded into the CNN as an extra layer at the end of the network. This results in real-time segmentation. To the best of our knowledge our method is the first attempt on general semantic image segmentation using CNN. All the previous papers were limited to few number of category of the images (e.g. PASCAL VOC). Experiments show that our segmentation algorithm outperform the state-of-the-art non-semantic segmentation methods by a large margin.