Toward an assisted indoor scene perception for blind people with image multilabeling strategies

作者：

Highlights：

• A novel coarse indoor scene description framework for blind people is introduced.

• Three image multilabeling implementation strategies are proposed.

• Experimental validation was conducted on different public indoor environments.

• Results qualify the approach for a near real time blind assistance technology.

摘要

•A novel coarse indoor scene description framework for blind people is introduced.•Three image multilabeling implementation strategies are proposed.•Experimental validation was conducted on different public indoor environments.•Results qualify the approach for a near real time blind assistance technology.

论文关键词：Assistive technologies,Bag of words,Blind people,Coarse indoor scene description,Image multilabeling,Principal component analysis,Scale invariant feature transform

论文评审过程：Available online 20 November 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.11.017