2 ms·
What an amazing idea :) They reproject the input images and run the low-res network multiple times. Then they use an approach similar to NeRF to merge the know
by fxtentacle 3y ago
What an amazing idea :)
They reproject the input images and run the low-res network multiple times. Then they use an approach similar to NeRF to merge the knowledge from those reprojected images into a super-resolution result.
So in a way, this is quite similar to how modern Pixel phones can take a burst of frames and merge them into a final image that has a higher resolution than the sensor. Except that they run useful AI processing in between and then do the super-resolution merge on the results.
- Grieverheart 3y agoAlso similar to temporal antialiasing https://en.wikipedia.org/wiki/Temporal_anti-aliasing https://en.wikipedia.org/wiki/Temporal_anti-aliasing .
- 3abiton 3y agoVery interesting, I am curious how do people reach that train of thought to a successful idea. So many great algorithms based on small twists.
- kreelman 3y agoIt is interesting indeed. One wonders if the researchers of this particular bit of work made it mandatory to go for walks at lunch and think about how their vision chunked/filtered the information it was receiving. Interesting that they "perturb" the image to get some noise involved. I'll need to read it over again.
- davecanderson 3y agoNature is such a good source of inspiration, the "perturb" approach reminded me of [fixational eye movement][1] but maybe that's only a clear link in retrospect. [1]: https://en.wikipedia.org/wiki/Fixation_(visual) https://en.wikipedia.org/wiki/Fixation_(visual)
- cryptonector 3y agoThis seems like it could have been inspired by how human vision works.
- foofie 3y ago[dead]
- jeffdaniels27x 3y agoPerhaps similar in some ways to how big cats' eyes reflect the light back from behind the retina (i.e. back through it for a second pass) to capture more light. I'm sure I heard that on a nature documentary ...