How close are we to solving vision?

Re:How close are we to solving vision? by KayEss on Sunday November 13, 2016 @11:24AM (#12944046)

Is each frame looked at separately? Given what is shown there seems to be no memory building context and pruning the options. Is that really hard to add?

[ Parent | on HN ]

Re:How close are we to solving vision? by omginternets on Sunday November 13, 2016 @02:17PM (#12945048)

There's something called "attentional neural networks" that attempt to do this. They tend to do very well in reading natural language, IIRC, but I've also seen them applied to video.

[ Parent | on HN ]