3 ms·
"the researchers are now working on the output imagery: instead of blobs, they want each moving target to be represented by a cross" -- That can't really be tha
by orjan 15y ago
"the researchers are now working on the output imagery: instead of blobs, they want each moving target to be represented by a cross" -- That can't really be that hard to implement. I'm sure they have bigger issues than this to deal with.
- mrsebastian 15y agoYeah, that's what I thought, but the press release suggests that it's harder than it sounds: http://www.mit.edu/newsoffice/2011/ll-seeing-through-walls-1018.html http://www.mit.edu/newsoffice/2011/ll-seeing-through-walls-1... “To understand the blobs requires a lot of extra training."
- bermanoid 15y agoI suspect the real problem is just that those images are so noisy - in the demo with two people moving around, for instance, it's almost impossible to tell at times whether there's one, two, or more blobs. Certainly it wouldn't be very reliable from a single frame, you'd have to use the movement information. Personally, I wouldn't waste my time on that, that's a rather straightforward image processing problem that could be made much easier by improving the output quality of the machine, which I imagine would be necessary anyways for this to see any real use in the field.
- Kliment 15y agoI believe that is clueless journalist speak for "they want to make an object classifier and tracker next, rather than just displaying raw data"
- shin_lao 15y agoEasy for your brain, hard for the machine.
- maaku 15y agoYeah, but it's not a new problem. See the Kinect, for example.