Seeing AI 2016 Prototype - A Microsoft research project

440K views · 10 years ago · demo · dated

A Microsoft research video of the 2016 prototype of Seeing AI, a smartphone app that describes the surrounding world to people who are blind or have low vision. It shows the app recognizing scenes, people and objects and speaking the results aloud.

HN people keep it as the reference for AI-assisted sight for the blind, citing it whenever new vision models or accessibility tools arrive.

8 comments from Hacker News

b
@bigtunacan2016 · in “Microsoft uses artificial intelligence to assist the blind”
Submitted the video to Hacker News.
RecommendsView on HN
l
@ljoshua2016 · in “Microsoft Cognitive Services: Introducing the Seeing AI App [video]”
Submitted the video to Hacker News.
RecommendsView on HN
_
@_isus2016 · in “Microsoft Cognitive Services: Introducing the Seeing AI App [video]”
Submitted the video to Hacker News.
RecommendsView on HN
s
@shagie2025 · in “Apple Releases Open Weights Video Model”
From a couple years ago... link ▶ this video link ... and that was 10 years ago. I'm curious for what it could do now.
s
@shagie2024 · in “Vision language models are blind”
As an aside... from 2016 this is what was a valid use case for a blind person with an app. Seeing AI 2016 Prototype - A Microsoft research project - ▶ this video link are the actual working apps. The version from 2016 I recall showing (pun not intended) to a coworker who had some significant vision impairments and he was really excited about what it could do back then. --- I still remain quite impressed with its ability to parse the picture and likely reason behind it link
CitesView on HN
s
@shagie2023 · in “Be My Eyes’ AI assistant starts rolling out”
▶ this video (At 45s in is the ideal) It's still not glass integrated - but the phone version has been out for a while ( link ) - previously discussed when it was released link (6 years ago, 133 comments)
s
@shagie2023 · in “Img2Prompt – Get prompts from stable diffusion generated images”
Something similar (image to text description) from a bit ago - Seeing AI app from Microsoft (2017) - link It's not prompt based intended to generate another one, but rather an accessibility tool. And some related videos: Seeing AI 2016 Prototype - A Microsoft research project - ▶ this video Seeing AI: Making the visual world more accessible - link
d
@dpflan2016 · in “A blind coder who makes Facebook accessible [video]”
Here is another example of improving accessibility, this time by Microsoft: ▶ this video