标题：Unifying Visual Localization and Scene Recognition for People With Visual Impairment             IEEE2020

INDEX TERMS： Visual place recognition, global image descriptor, scene classification,assistivenavigation

针对”Where am I?“and ”What are the surroundings?“，主要包含场景识别，场景描述和视觉定位

批：整体应该是做的一个搜索的工作，帮助不大

-————————————————————————————————————————————————————————————

标题："Person, Shoes, Tree. Is the Person Naked?" What People  with Vision Impairments Want in Image Descriptions

Author  Keywords ： Image captions, alt text, accessibility, visual impairment

主要针对的是图像描述偏好如何根据遇到数字图像的来源和周围环境而变化

-————————————————————————————————————————————————————————————

标题：Going Beyond One-Size-Fits-All Image Descriptions to Satisfy the Information Wants of People Who are Blind or Have Low Vision

Additional Key Words and Phrases: image description, image caption, alt text, blind, low vision, visual impairment, scenarios, minimum viable description, context awar

把描述放到场景（scenarios）里，或者说场景基的图像描述。1.从一刀切（one-size-fits-all）描述到包含背景感知描述 2.大场景下的最小可行描述

-————————————————————————————————————————————————————————————

标题：Assistive Technology for the Blind（在312页）

Keywords YOLO· LSTM· InceptionV3· Pyttsx3· Image captioning

整体技术路线是相机--image caption--转换音频输出

批：优先看

-————————————————————————————————————————————————————————————

标题：Assisted ?nvironment ?omprehension using a Mobile Terminal Application

keywords—visually impaired; scene captioning; deep neural networks;embedding processing;. 

模型部署在web，app访问服务器，服务器返回图像描述。文章大概侧重在部署把，没看到什么很新的东西。

-————————————————————————————————————————————————————————————

标题：Wearable Visually Assistive Device for Blind People to Appreciate Real-world Scene and Screen Image        VCIP2020，很短

Index  Terms— blindness, assistive system , vision disorder 

技术路线：深度相机--image caption--转换音频输出

系统分为四个模块： 1.目标检测 2.image caption 3.测距模块 4.交互模块

-————————————————————————————————————————————————————————————

未完待续...