News for hackers. Stuff that matters.
Re:Deep Learning in a Single File for Smart Devices by joshvm on Thursday November 12, 2015 @07:56AM (#10552797)
I saw a demo of a deep convnet (I assume) running on a tablet at a conference recently. While it's limited to the vocabulary that the net is trained on, it's seriously impressive seeing this stuff work in realtime.

One of the other speakers was giving a tech demo using his webcam. He was looking around for a mug to demonstrate that the classification was good and could work quickly. In the meantime the camera was looking behind him on stage and correctly classified the image as "theatre curtains". It was particularly cool because image processing results are often cherrypicked to show optimal performance and you learn to be skeptical.

Re:Deep Learning in a Single File for Smart Devices by zk00006 on Thursday November 12, 2015 @08:44AM (#10553052)
Thumb up for Qualcomm demo and Andrea Vedaldi presentation of MatConvNet presented at ICVSS 2015 summer school (not a conference). Both pieces made similar impression on me.
Re:Deep Learning in a Single File for Smart Devices by joshvm on Friday November 13, 2015 @12:25AM (#10558170)
Yep I was thinking of Vedaldi's presentation - my bad, it was indeed a summer school. They all blur into one these days!
Re:Deep Learning in a Single File for Smart Devices by oxplot on Thursday November 12, 2015 @11:58AM (#10554348)
Any links?