computer vision
Teaching a machine to interpret images: recognising, classifying and organising what is in a picture.
FilterEveryone, all time21 from 12 people
- FL
Fei-Fei Li quoted
Sequoia Professor of Computer Science at Stanford
Their wordsIt's the first time we have a unification of pixel generation and pixel reconstruction. In the world of computer vision this field has been around for more than half a century.
↗Why Fei-Fei Li Is Betting on Spatial Intelligenceyoutube.com 1st of 13 in this recording
- 6 weeks earlier
- JH
Jensen Huang quoted
Co-founder and CEO of NVIDIA since 1993
Their wordsthe the breakthrough for us was realizing that AlexNet was not AlexNet. That AlexNet was an approach with deep deep learning that allows you to learn any function.
↗Jensen Huang: The Mindset That Built NVIDIAyoutube.com 5th of 20 in this recording
- 8 months earlier
- EY
Eugene Yan quoted
Member of technical staff at Anthropic
Their wordsAnd human annotators can miss as many as 50% of the defects due to fatigue after looking at hundreds of samples.
↗Product Evals in Three Simple Stepseugeneyan.com 2nd of 3 in this piece
- 5 weeks earlier
- AK
Andrej Karpathy quoted
Founding member of OpenAI and former director of AI at Tesla
Their wordsso a good example recently was um Jeff Hinton's prediction that radiologists would not be a job anymore and this turned out to be very wrong in a bunch of ways right so radiologists are alive and well and growing even though computer vision is really really good at recognizing all the different things that they have to recognize
↗Andrej Karpathy — “We’re summoning ghosts, not building animals”youtube.com 19th of 30 in this recording
- 3 weeks earlier
- AK
Andrej Karpathy quoted
Their wordsExpectation: rapid progress in image recognition AI will delete radiology jobs (e.g. as famously predicted by Geoff Hinton now almost a decade ago). Reality: radiology is doing great and is growing.
- 13 days earlier
- SL
Sergey Levine quoted
Professor of computer science at UC Berkeley
Their wordsWhereas with text, it's already sort of been abstract into those bits that we as humans care about. So the representations are already there and they're not just good representations. They actually like focus in on what really matters.
↗Fully autonomous robots are much closer than you think – Sergey Levineyoutube.com 12th of 24 in this recording
- SL
Sergey Levine quoted
Their wordsUh and its perception is in service to fulfilling that purpose. And that is like a really great uh focusing factor. We know that for people this really matters. Like literally what you see is affected by what you're trying to do.
↗Fully autonomous robots are much closer than you think – Sergey Levineyoutube.com 13th of 24 in this recording
- 17 months earlier
- LW
Lilian Weng quoted
Machine-learning researcher
Their wordsIn comparison to text or images, it is more difficult to collect large amounts of high-quality, high-dimensional video data, let along text-video pairs.
↗Diffusion Models for Video Generationlilianweng.github.io 2nd of 2 in this piece
- 6 months earlier
- CH
Chip Huyen quoted
Engineer and writer on machine-learning systems
Their wordsImage is perhaps the most versatile format for model inputs, as it can be used to represent text, tabular data, audio, and to some extent, videos. There's also so much more visual data than text data.
↗Multimodality and Large Multimodal Models (LMMs)huyenchip.com 2nd of 3 in this piece
- 12 months earlier
- SA
Scott Alexander quoted
Essayist; writes Astral Codex Ten
Our reading · no longer heldThe three-year bet about whether AI image models could handle compositionality was won in three months.
No longer holdsI retract my claim to have won and will continue to see how AI progress advances over the next three years.
SaidI retract my claim to have won and will continue to see how AI progress advances over the next three years.
- 4 weeks earlier
- SA
Scott Alexander quoted
Their wordsIn fact, not only did they reach this level in three months, but probably they were at this level before the bet was even made
↗I Won My Three Year AI Progress Bet In Three Monthsastralcodexten.substack.com
- 17 months earlier
- GW
Gwern quoted
Pseudonymous essayist
Our readingUnsupervised learning trained on unfiltered, internet-scale image data now genuinely works.
Their wordsanother reminder that unsupervised learning is really working!
↗March 2021 Gwern.net Newslettergwern.substack.com 1st of 3 in this piece
- 5 years earlier
- KH
Kaiming He quoted
Associate professor at MIT and first author of the residual networks paper
Their wordsThe depth of representations is of central importance for many visual recognition tasks.
↗He, Zhang, Ren & Sun, "Deep Residual Learning for Image Recognition" (arXiv)arxiv.org All 8 from this paper
- JS
Jian Sun quoted
Computer vision researcher
Their wordsThe depth of representations is of central importance for many visual recognition tasks.
↗He, Zhang, Ren & Sun, "Deep Residual Learning for Image Recognition" (arXiv)arxiv.org All 8 from this paper
- 7 years earlier
- FL
Fei-Fei Li quoted
Their wordsThe explosion of image data on the Internet has the potential to foster more sophisticated and robust models and algorithms to index, retrieve, organize and interact with images and multimedia data.
↗ImageNet: A large-scale hierarchical image database (with 5 co-authors)doi.org All 4 from this paper
- FL
Fei-Fei Li quoted
Our readingHow large-scale internet image data can be harnessed and organized remains a critical problem.
Their wordsBut exactly how such data can be harnessed and organized remains a critical problem.
↗ImageNet: A large-scale hierarchical image database (with 5 co-authors)doi.org All 4 from this paper
- FL
Fei-Fei Li quoted
Our readingImageNet is much larger in scale and diversity and more accurate than existing image datasets.
Their wordsWe show that ImageNet is much larger in scale and diversity and much more accurate than the current image datasets.
↗ImageNet: A large-scale hierarchical image database (with 5 co-authors)doi.org All 4 from this paper
- FL
Fei-Fei Li quoted
Their wordsWe hope that the scale, accuracy, diversity and hierarchical structure of ImageNet can offer unparalleled opportunities to researchers in the computer vision community and beyond.
↗ImageNet: A large-scale hierarchical image database (with 5 co-authors)doi.org All 4 from this paper
- 11 years earlier
- YB
Yoshua Bengio quoted
Deep-learning researcher and Turing Award laureate
Their wordsGiven an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing.
↗Gradient-based learning applied to document recognition (with 3 co-authors)doi.org All 4 from this paper
- YB
Yoshua Bengio quoted
Their wordsConvolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques.
↗Gradient-based learning applied to document recognition (with 3 co-authors)doi.org All 4 from this paper
- YB
Yoshua Bengio quoted
Their wordsA new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure.
↗Gradient-based learning applied to document recognition (with 3 co-authors)doi.org All 4 from this paper