HackerTrans
TopNewTrendsCommentsPastAskShowJobs

skandium

no profile record

Submissions

The changing goalposts of AGI and timelines

mlumiste.com
404 points·by skandium·hace 4 meses·388 comments

The Macroeconomics of Agentic AI: Are We the Peasant or the Horse?

mlumiste.com
1 points·by skandium·hace 5 meses·0 comments

Animating old family photos for $0.36 each

mlumiste.com
2 points·by skandium·hace 6 meses·0 comments

Perceptually lossless (talking head) video compression at 22kbit/s

mlumiste.com
225 points·by skandium·hace 2 años·140 comments

Compressing Images with Neural Networks

mlumiste.com
171 points·by skandium·hace 2 años·69 comments

Quarterly planning is a noisy optimisation problem

mlumiste.com
3 points·by skandium·hace 3 años·0 comments

comments

skandium
·el año pasado·discuss
One of the great tragedies of the world is that while he is arguably the philanthropist with the highest positive impact in human history, a significant part of the population seems to still think he is the literal Antichrist.
skandium
·hace 2 años·discuss
Well, this isn't probably a problem with the model, but the source frame having wrong eye gaze. Besides, perceptually lossless need not be defined in a side-by-side comparison context. If you were only viewing the right hand side video, how could you tell the eye gaze is off? The point was more on that the movement looks natural, unlike almost all neural avatars up to this year.
skandium
·hace 2 años·discuss
This is my field as well, although I come from the neural network angle.

Learned video codecs definitely do look promising: Microsoft's DCVC-FM (https://github.com/microsoft/DCVC) beats H.267 in BD-rate. Another benefit of the learned approach is being able to run on soon commodity NPUs, without special hardware accommodation requirements.

In the CLIC challenge, hybrid codecs (traditional + learned components) are so far the best, so that has been a letdown for pure end to end learned codecs, agree. But something like H.267 is currently not cheap to run either.
skandium
·hace 2 años·discuss
If you look at the winners of the Hutter prize, or especially the Large Text Compression Benchmark, then almost every approach uses some kind of machine learning approach for the adaptive probability model and then either arithmetic coding or rANS to losslessly encode it.

This is intuitive, as the competition organisers say: compression is prediction.
skandium
·hace 2 años·discuss
Exactly, in practice the alternatives are either blocky artifacts (JPEG and most other traditional codecs), blurring everything (learned codecs optimised for MSE) or "hallucinating" patterns when using models like GANs. However, in practice even the generative side of compression models is evaluated against the original image rather than only output quality, so the outputs tend to be passable.

To see what a lossy generator hallucinating patterns means in practice, I recommend viewing HiFiC vs original here: https://hific.github.io/