Sergey Korolev

ML/CV/AR @ Snap Inc.

[Sergey Korolev @libfun_sk]

Very happy to bring generative 4D to the hands of our users!

This was a great collab with Snap Research and would not have been possible without @wang12_gordon and @ashmrz10’s paper that was published this year on NeurIPS!

https://t.co/RlGwmzNGea

[Lens Studio @LensStudioDev]

A gift to you from the GenAI Suite this season 🎁

Let’s see what's inside:
🌠 Animated Selfie Attachments
Generate animated 3D assets for your next selfie Lens.
❄️ New Enhanced Model
Take your Style and Face Generations to an entirely new level with the new Enhanced model, now added to both features.
⛄ Turn Heads with Updated Head Morph
Now get better quality outputs for characters that live up to your imagination.

Happy Holidays from Lens Studio! https://t.co/17AyvE5wsn

[Vladislav Shakhray @shakhrayv]

Really excited to share what my team’s been cooking at Snap!

Selfie Attachments let you dream up anything and instantly bring it into AR.
It’s a big step toward the future of Generative AR, and I’m excited to see where it takes us next.

[Ashkan Mirzaei @ashmrz10]

[1/9] 🚀 We introduce 4Real-Video-V2, a method that can generate 4D scenes from a simple text prompt, viewable from any angle at any moment in time. It’s fast, photorealistic, and works on full scenes. Here's how it works and why it matters. 👇

https://t.co/vr9Wnvfx0s

[Kate Deyneka @katedeyneka]

Our team released the new Lens Studio where you can create Snapchat Lenses directly from your phone or browser! I’m particularly excited about GenAI features there. Congrats to the team!

iOS: https://t.co/A3eg48BM6g
Web: https://t.co/GGIoMaZPWD

[Sergey Korolev @libfun_sk]

SnapML Kit is out, with a new feature allowing to generate rigged and animated high quality head models just from a text prompt to unlock creative lenses in Snapchat. Try it out at https://t.co/cn2UfM7oUz
#3Dgeneration

[Sergey Korolev @libfun_sk]

Snap has launched SnapML Kit today. New tool that allows anyone to generate 3D assets with just a text prompt or an image. Check it out at https://t.co/cn2UfM7oUz
#3Dgeneration

[Sergey Tulyakov @SergeyTulyakov]

We have started accepting Snap Research 2022 Fellowship applications! We will award up to 10 fellowships. If you're doing highly impactful research relevant to Snap, we encourage you to apply. The deadline is Oct 10. Apply here:
https://t.co/jIMySVzqYi

[Yury Kashnitsky @ykashnitsky]

Topic 7 of the Open Machine Learning Course by @odsai_en and @libfun_sk is dedicated to unsupervised learning, namely, to Principal Component Analysis and clustering. https://t.co/N2UV6Lc9tw #mlcourse_open (credit for image to CS231n)

[Sergey Korolev @libfun_sk]

I just published “Open Machine Learning Course. Topic 7. Unsupervised learning: PCA and clustering” https://t.co/8fvQAj6L6Z

[Andrej Karpathy @karpathy]

I've been using PyTorch a few months now and I've never felt better. I have more energy. My skin is clearer. My eye sight has improved.

[Ian Goodfellow @goodfellow_ian]

@hugo_larochelle @m_aggrey @ylecun

[tom white @dribnet]

what models learn: so many strange surprises in the imagenet dataset. here's training image 553/1300 for category: "garbage truck"

[Andrej Karpathy @karpathy]

New post on placing AlphaGo in context of AI research https://t.co/tI3VkUGYsT trying to mix in a pinch of low-level depth to popsci/PR

[Andrey Ustyuzhanin @AnaderiRu]

Summer school on ML for HEP - Reading, UK 17-23 Jul. Expect cool speakers, ML theory and extensive hands-ons https://t.co/d4ZZLFRvhs

[NeurIPS Conference @NeurIPSConf]

Videos from the 2016 Tutorials and Conference are available at https://t.co/MQAXlggjbX. Workshop videos will be posted at a later date.

[Jack Clark @jackclarkSF]

OH at lunch at OpenAI: "NIPS is becoming like Burning Man"

[Artëm Sobolev @art_sobolev]

A collection of summaries / reviews / highlights / etc of #nips2016 https://t.co/lFQG3Rlxzv

[Damon Civin @DCivin]

A sign of a lot of generative models at #nips2016 #gan #vae

[François Chollet @fchollet]

The most surprising thing from NIPS so far is the continued prominence of MNIST in experiments.