After spending hours doing n^2 manual searching and copying 50 photos from 500.. I wish digital photography was a computing platform. Why can’t I write quick minimal programs to automate renames? Why can’t I do offline face tagging? Why can’t I query my photos on arbitrary props?
Computer vision is hard, and systems for it are ad hoc. Need an integrated programming model (think interactive, probabilistic SQL) to make it sufficiently easy for an end user. This is where my current research is heading.
-
-
Yeah, we still don’t have a computer vision language not using Fortran or another super old language IIRC. I read some of your research and found it intriguing, I wonder if there are examples of non-SQL languages for interactive queuing of rich objects like photos.
-
Neat idea, esp because ML has come so far but value is mostly locked in non-open systems like Google Photos. should be totally possible to implement a language like “CONTAINS dog IN left top quadrant OR cat”
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
cognitive psychology. PhD