Artsiom Sanakoyeu @artsiom_s Twitter profile | Pikagi

Pikagi

Artsiom Sanakoyeu

@artsiom_s

3,751

Followers

662

Following

383

Media

2,774

Statuses

Staff Research Scientist @Meta Generative AI PhD in Computer Vision @ Heidelberg University, @Kaggle Competitions Master (Top-50 worldwide)

Zürich, Switzerland

https://t.co/4QwKgafjpR

Joined December 2016

Don't wanna be here? Send us removal request.

Pinned Tweet

@artsiom_s

Artsiom Sanakoyeu

4 months

Happy to announce Imagine Flash, which is a real-time image synthesis! Watch in real time as the image evolves with each character you type! I'm proud to be leading the Flash project with my teammates - it's incredibly rewarding to witness the transformation of a quick demo I

3

7

79

Last Seen Profiles

@dwarlisst

@radh_al52040

@Yas_words

@LondonOnDaTrack

@uem4uk

@Drsaudiman

@Jonny_Dutch

@JohnFMiller86

@bokeplokalmalam

@LaWalletOk

@DegenHQ_

@FakKrba

@CodyZeller

@JRCariani

@bertpsch

@KUONN_931

@GuardiaCrema

@KadriGursel

@Katerationopia

@KILLYOURCELLMAX

@FVG_Hunter

@Goldensoph67668

@Idontcarepal5

@FootyGoth

@Katiyaaahh

@PlaySparc

@Jennoption

@Kari_Squared

@henryakang98555

@HDescendimiento

@Aris_K_182

@JingBai_ZhangBi

@J5daigada

@GoMHSIndians

@masor_konya

@Ramitagram1

@artsiom_s

Artsiom Sanakoyeu

3 years

The shortest guide for pytorch training on GPUs

Tweet media one

16

286

2K

@artsiom_s

Artsiom Sanakoyeu

3 years

StyleGAN3 is out! Here is Colab:

4

142

911

@artsiom_s

Artsiom Sanakoyeu

4 months

⚡️SD3-Turbo: Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation Following Stable Diffusion 3, my ex-colleagues have published a preprint on SD3 distillation using 4-step, while maintaining quality. The new method – Latent Adversarial Diffusion

Tweet media one

Tweet media two

Tweet media three

14

91

428

@artsiom_s

Artsiom Sanakoyeu

3 years

You don't need EfficientNets. Simple tricks make ResNets better and faster than EfficientNets Revisiting ResNets: Improved Training and Scaling Strategies 🤙

Tweet media one

Tweet media two

Tweet media three

10

99

413

@artsiom_s

Artsiom Sanakoyeu

3 years

I'm happy to announce that yesterday I defended my PhD in Computer Vision!!!🥳🍾

Tweet media one

15

3

407

@artsiom_s

Artsiom Sanakoyeu

3 years

Self-supervised Learning for Medical images Due to fixed imaging procedures, medical images like X-ray or CT scans are usually well aligned. This gives an opportunity to utilize such an alignment to automatically mine similar pairs of images for training

Tweet media one

1

63

340

@artsiom_s

Artsiom Sanakoyeu

3 years

Swin Transformer: New SOTA backbone for Computer Vision🔥 👉 What? New vision Transformer architecture called Swin Transformer that can serve as a backbone in computer vision instead of CNNs. 📝 ⚒ Code (soon) Thread 👇

Tweet media one

3

76

335

@artsiom_s

Artsiom Sanakoyeu

1 year

We have released the code and weights for our #CVPR2023 paper "Avatars Grow Legs: Generating Smooth Human Motion from Sparse Tracking Inputs with Diffusion Model"! code: abs: project: The demo is below:

4

53

313

@artsiom_s

Artsiom Sanakoyeu

4 years

Recently, me and my team secured 3rd place ($6k prize) at Kaggle competition "Lyft Motion Prediction for Autonomous Vehicles" 🛠️ Code Solution:

Tweet media one

3

38

282

@artsiom_s

Artsiom Sanakoyeu

3 years

The Rendering equation explained. Useful for understanding FastNeRF.

Tweet media one

3

20

254

@artsiom_s

Artsiom Sanakoyeu

5 months

Staff Research Scientist: Personal Update I have some exciting news that I'd like to share with you! On Monday, I was promoted to E6, which means I am now a Staff Research Scientist at Meta GenAI. This was made possible thanks to the significant impact and scope of a Generative

Tweet media one

21

4

230

@artsiom_s

Artsiom Sanakoyeu

3 years

One of the last articles from interactive web-journal . "A Gentle Introduction to Graph Neural Networks"

Tweet media one

0

39

205

@artsiom_s

Artsiom Sanakoyeu

1 year

Lol. Dude it's not the model that takes 100 Mb, but an extra thing that they train on top of 1B parameter model! Don't distribute fake news

@javilopen

Javi Lopez ⛩️

1 year

🔴 PERFUSION: a generative AI model from NVIDIA that fits on a floppy disk 💾 It takes up just 100KB. Yes, you heard it right, much less than any picture you take with your mobile phone! Why is this revolutionary and can change everything? I'll tell you 🧵👇

Tweet media one

38

273

2K

8

12

186

@artsiom_s

Artsiom Sanakoyeu

1 year

Come to see our #CVPR2023 poster "Avatars Grow Legs: Generating Smooth Human Motion from Sparse Tracking Inputs with Diffusion Model". Learn how to synthesize full body motion based on head and wrists only! webpage: Today at 10:30-12:30, poster #46 .

0

34

169

@artsiom_s

Artsiom Sanakoyeu

3 years

Neural 3D Video Synthesis NERF-like model generates frames conditioned on position, view direction and time-variant latent code. When it gets faster, it will enable mind-blowing applications! 📝 🌐

4

41

169

@artsiom_s

Artsiom Sanakoyeu

3 years

Really nice introduction into hyped Diffusion Models by @lilianweng . With (almost) all necessary theory packed inside.

0

20

154

@artsiom_s

Artsiom Sanakoyeu

3 years

⚔️ FastNeRF vs NeX ⚔️ Smart ideas do not come in the only head. FastNeRF has the same idea as in NeX, but a bit different implementation. Which one is Faster? Nex FastNeRF To learn about differences between the two -> thread 👇

Tweet media one

2

31

150

@artsiom_s

Artsiom Sanakoyeu

3 years

NeX: Real-time View Synthesis with Neural Basis Expansion An amazing new approach to novel view synthesis a combination of multiplane image (MPI) and neural basis expansion (NeRF-like nets). It can reproduce spectacular complex view-dependent effects 🌐

1

29

140

@artsiom_s

Artsiom Sanakoyeu

3 years

Check out our new #CVPR21 paper! Discovering Relationships between Object Categories via Universal Canonical Maps In collaboration with FAIR ( @NataliaNeverova , P. Labatut, @davnov134 and A. Vedaldi) 🌐 ▶️ 📝

2

23

134

@artsiom_s

Artsiom Sanakoyeu

4 years

StyleGAN2 for transferring garments between different poses and body shapes. The results are pretty neat! Virtul try-on is coming soon folks! 🌎Project page: 🧥Interactive example:

4

28

121

@artsiom_s

Artsiom Sanakoyeu

4 years

A post showing some magic python visualization abilities.

Tweet media one

3

18

101

@artsiom_s

Artsiom Sanakoyeu

1 year

Our paper "Avatars Grow Legs" (CVPR 2023) is out! TL; DR: Fast Diffusion models to generate full body motions based on head and hands tracking inputs, Will release the code in a few days.

@_akhaliq

AK

1 year

Avatars Grow Legs: Generating Smooth Human Motion from Sparse Tracking Inputs with Diffusion Model abs: project page:

6

67

322

2

16

100

@artsiom_s

Artsiom Sanakoyeu

3 years

To learn more about our 3rd place solution for the @Kaggle @LyftLevel5Motion "Lyft Prediction for Autonomous Vehicles competition" read my blogpost:

Tweet media one

0

18

99

@artsiom_s

Artsiom Sanakoyeu

3 years

My new video on self-supervised representation learning (also easy to understand for beginners). I explain CliqueCNN which builds compact cliques for classification as a pretext task and I discuss other self-supervised learning approaches. @itsbautistam

Tweet media one

3

19

95

@artsiom_s

Artsiom Sanakoyeu

5 years

Last week I gave a talk at Heidelberg SIAM chapter "Identification of Humpback Whales using Deep Metric Learning". I talked about our recent CVPR'19 paper and about Humpback Whale Identification challenge at @kaggle . Slides:

Tweet media one

0

36

89

@artsiom_s

Artsiom Sanakoyeu

3 years

StyleGAN2-ADA train on cute Corgi images. Looks amazing! Around 130k 1024x1024 images used. 🌀Colab 🛠️ Code

4

25

96

@artsiom_s

Artsiom Sanakoyeu

10 months

Since joining Meta GenAI, our team focused on speed advancements in image synthesis. Exciting news!🚀 We've unlocked high-quality image synthesis in just ~5 sec! MZ showcased our progress at Meta Connect. Try it out with the /imagine command in our AI chatbot in FB, IG or WA

8

10

92

@artsiom_s

Artsiom Sanakoyeu

3 years

Hiring interns for our team in Reality Labs Zurich! We are looking for PhD students with a strong research background, proven by publications in top-tier venues. The primary goal of the internship is to submit a paper to CVPR 2023. Details in the thread⬇️

Tweet media one

2

16

93

@artsiom_s

Artsiom Sanakoyeu

3 years

A blogpost from Apple summarizing their research on generative model for scene level radiance fields (GSN), ICCV 2021

Learning to Generate Radiance Fields of Indoor Scenes

People have an innate capability to understand the 3D visual world and make predictions about how the world could look from different points…

machinelearning.apple.com

2

23

90

@artsiom_s

Artsiom Sanakoyeu

3 years

Cool work from @facebookai ! It can generate an image of input text in any style provided an example of reference style. Architecture loosk similar to StyleGAN, but instead of noise, every nomalization layer is conditioned on the encoded style vector.

1

13

78

@artsiom_s

Artsiom Sanakoyeu

4 years

Check out our new #CVPR20 paper on Transferring DensePose to Animals In collaboration with FAIR (V. khalidov, A. Vedaldi and @NataliaNeverova ) 🌐 ▶️ 📝

Tweet media one

1

19

71

@artsiom_s

Artsiom Sanakoyeu

5 years

We willl be presenting our work "Divide and Conquer the Embedding Space for Metric Learning" at #CVPR2019 on Tuesday 18th: Poster 24 at 10:15. Authors: Me, Vadim Tschernezki, Uta Büchler ( @uta0590 ) and Björn Ommer. Paper and Code:

Tweet media one

0

25

68

@artsiom_s

Artsiom Sanakoyeu

4 years

I will present our #CVPR2020 paper on Transferring Dense Pose to Animals Today at 10am PDT / 7PM CET. Join Q&A 🌐 📝

0

24

69

@artsiom_s

Artsiom Sanakoyeu

3 years

Barlow Twins: Self-Supervised Learning via Redundancy Reduction New self-supervised learning loss: compute cross-correlation matrix between the features of two distorted versions of a sample and make it close to the identity. 🛠️

Tweet media one

2

17

65

@artsiom_s

Artsiom Sanakoyeu

3 years

🔥New DALL-E? Paint by Word 🔥 Edit a generated image by painting a mask atany location of the image and specifying any text description. Or generate a full image just based on textual input. 📝 1/

Tweet media one

2

11

65

@artsiom_s

Artsiom Sanakoyeu

3 years

CvT: Introducing Convolutions to Vision Transformers🔥 SOTA ImageNet Results (almost) Inject Inductive biases of CNNs (i.e. shift, scale, and distortion invariance) to the ViT architecture while maintaining the flexibility of Transformers. 📝 Thread👇

Tweet media one

1

19

64

@artsiom_s

Artsiom Sanakoyeu

4 years

Dense pose for animal classes with transfer learning @facebookai blog post about our #CVPR20 paper. 🌐 Blog 📝 Paper

Tweet media one

1

17

64

@artsiom_s

Artsiom Sanakoyeu

5 years

Our paper on training with pseudo-labels for semantic segmentation, GCPR 2019. Semi-Supervised Segmentation of Salt Bodies in Seismic Images: SOTA (1st place) at TGS Salt Identification Challenge. 🌐 📝 #kaggle #TGS2019 #GCPR19

Tweet media one

Tweet media two

1

20

62

@artsiom_s

Artsiom Sanakoyeu

2 years

In 1 hour at #NeurIPS2022 I will be presenting VisCo-Grids, a grid-based surface reconstruction method incorporating Viscosity and Coarea priors. Joint work at @MetaAI with @AlbertPumarola , @YarivLior , @alitabet and @lipmanya Details in the thread 🧵

1

15

63

@artsiom_s

Artsiom Sanakoyeu

3 years

New video on my YouTube channel! In this video, I explain VectorNet - a method for future motion prediction based on a vectorized representation of the scene instead of RGB images. 🎬

1

15

62

@artsiom_s

Artsiom Sanakoyeu

3 years

I'm happy to announce that our team (me, @KonevSteven , K. Brodt) was awarded 3rd place within the Waymo Motion Prediction Challenge 🥳 Task: predict trajectories of the agents for 8 seconds into the future. 📜Technical report We also released our code ↓

Tweet media one

3

11

60

@artsiom_s

Artsiom Sanakoyeu

3 years

How to easily edit and compose images like in Photoshop using GANs🔥 ❓What? Given an incomplete image or a collage of images, generate a realistic image 📌How? 1.Train a regressor to predict StyleGAN latent code even from incomplete image 2.Embedd collage and send it to GAN

Tweet media one

3

11

60

@artsiom_s

Artsiom Sanakoyeu

10 months

Presenting our work "Re-ReND: Real-time Rendering of NeRFs across Devices" #ICCV23 We show how to bake a NeRF on a mesh with rich view-dependent textures to allow rendering 100-1000 FPS on different devices without loss of quality. Visit our poster: ID: 3760 Foyer Sud"- 140

1

8

61

@artsiom_s

Artsiom Sanakoyeu

3 years

Self-supervised learning: The dark matter of intelligence Blog post by @ylecun and @ishan_ - well-known experts in self-supervised learning at FAIR. 0/5 They talk about: - Self-supervised learning as a paradigm in general ...

Tweet media one

2

20

60

@artsiom_s

Artsiom Sanakoyeu

4 years

I wrote a blog post which briefly explains the SMAL model for fitting 3D shapes of animals to RGB images paper. Based on paper “3D Menagerie: Modeling the 3D Shape and Pose of Animal”, CVPR 2017 @silvia_zuffi @Michael_J_Black 🌐

Tweet media one

2

13

58

@artsiom_s

Artsiom Sanakoyeu

4 years

Played around with neural rendering. Here is the result of COLMAP of 100 photos + rendering the 3D points using Neural Point-Based Graphics

1

11

57

@artsiom_s

Artsiom Sanakoyeu

3 years

StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery 🔥 Use CLIP model in order to navigate image editing in StyleGAN by text queries. 📝Paper ⚙️ code Thread 👇

Tweet media one

Tweet media two

2

9

53

@artsiom_s

Artsiom Sanakoyeu

4 years

Papers from A. Efros are always the top 🔝 One of the favourite papers I've recently read: Space-Time Correspondence as a Contrastive Random Walk * Tracking w/o supervision using random walk between image patches. 🌐

Tweet media one

Tweet media two

Tweet media three

0

6

51

@artsiom_s

Artsiom Sanakoyeu

2 years

Happy to share that we got 1/1 #NeurIPS2022 papers acepted this year from our small team in Reality Labs Zurich! Working on the camera ready and will upload it to arXive soon. Small spoiler: it's on learning implicit 3D shape representations for shape reconstruction.

1

0

51

@artsiom_s

Artsiom Sanakoyeu

3 years

New Blogpost! Google showed why most of the recent Transformer Modifications Fail To Transfer Across Implementations and Applications.

0

9

49

@artsiom_s

Artsiom Sanakoyeu

3 years

Some cool results from VQGAN+CLIP experiments 1. "Holy war against capitalism" 2. "Polygonal fast food" 3. "Minecraft Starcraft" 4. "Modern cubist painting" 🎩Colab:

Tweet media one

Tweet media two

Tweet media three

Tweet media four

3

11

48

@artsiom_s

Artsiom Sanakoyeu

5 years

Our team (me, @ppleskov and @shakhrayv ) finished 10th (out of 2131 teams) in Humpback Whale Identification challenge on @kaggle . Special thanks to @odsai_en community for fruitful discussions!

Tweet media one

6

14

49

@artsiom_s

Artsiom Sanakoyeu

3 years

Transformer Meets Tracker: Exploiting Temporal Context for Robust Visual Tracking 📝Paper: 🛠️

Tweet media one

0

11

46

@artsiom_s

Artsiom Sanakoyeu

4 months

🔥Fresh drop - Mixtral-8x22B! As usual, @MistralAI stays true to their style by simply leaving a magnet link to a torrent with the weights of their new model. Nice trolling! The new model is a Mixture of Experts Mixtral-8x22B: - Model size: 262 GB (I assume the weights are in

Tweet media one

3

4

40

@artsiom_s

Artsiom Sanakoyeu

1 year

I'm at #CVPR2023 . Dm me if you want to catch up!

Tweet media one

2

3

40

@artsiom_s

Artsiom Sanakoyeu

4 months

Mark @finkd is talking about our Imagine Flash right here. I won't lie, it feels really good when your CEO speaks about your work this way 🙂

2

1

39

@artsiom_s

Artsiom Sanakoyeu

4 years

My youtube video explaning HOW to EARN $6000 By WINNING A KAGGLE AUTONOMOUS DRIVING COMPETITION. The video is on our 3rd place solution for the @Kaggle @LyftLevel5 Motion Prediction for Autonomous Vehicles competition. 🛠️ 🎬

Tweet card media

GitHub - asanakoy/kaggle-lyft-motion-prediction-av: The 3rd place solution for competition "Lyft...

The 3rd place solution for competition "Lyft Motion Prediction for Autonomous Vehicles" at Kaggle - asanakoy/kaggle-lyft-motion-prediction-av

0

13

39

@artsiom_s

Artsiom Sanakoyeu

5 years

Best Paper award #iccv19 : SinGAN I really liked their results on the task of Super resolution!

Tweet media one

Tweet media two

Tweet media three

Tweet media four

2

5

40

@artsiom_s

Artsiom Sanakoyeu

6 years

Source code and pretrained models for our paper A Style-Aware Content Loss for Real-time HD Style Transfer are on GitHub! Website: #ECCV2018 #ECCV

Tweet media one

Tweet media two

1

14

40

@artsiom_s

Artsiom Sanakoyeu

6 years

Our paper was accepted as oral at ECCV 2018! "A Style-Aware Content Loss for Real-time HD Style Transfer" Artsiom Sanakoyeu*, Dmytro Kotovenko*, Sabine Lang ( @lang254 ) , Björn Ommer Project page: Source code is coming soon.

Tweet media one

Tweet media two

Tweet media three

0

16

38

@artsiom_s

Artsiom Sanakoyeu

4 years

I'm delighted to share that I was selected as an outstanding reviewer at NeurIPS for the second time in a row! #neurips2020

Tweet media one

2

0

37

@artsiom_s

Artsiom Sanakoyeu

4 years

Our #CVPR2020 paper on DensePose for animals got covered in the weekly AI newsletter of , @AndrewYNg 's AI education startup

1

9

37

@artsiom_s

Artsiom Sanakoyeu

1 year

A few weeks ago, Mark announced the creation of a new organization within #Meta - GenAI, focusing solely on Generative AI. Our team has left Reality Labs & joined the new org. Thrilled as I've been working on diffusion models for the past year - now full steam ahead! 🚀 #GenAI

Tweet media one

1

1

36

@artsiom_s

Artsiom Sanakoyeu

2 years

We're looking for talented PhD interns to join our team at Meta Reality Labs in Zurich. Our focus is on 3D human motion synthesis & tracking for AR/VR, and we're offering the chance to work on cutting-edge technology like generative models (diffusion, VAEs) for motion synthesis

Tweet media one

4

3

35

@artsiom_s

Artsiom Sanakoyeu

3 years

My PhD thesis: "Visual Representation Learning with Limited Supervision"

Tweet media one

0

4

35

@artsiom_s

Artsiom Sanakoyeu

3 years

I was promoted to the rank of Expert Reviewer by @icmlconf . This is nice and gives some extra motivation to keep the high quality of reviews!

Tweet media one

0

0

34

@artsiom_s

Artsiom Sanakoyeu

4 years

My blog post on how to design a container with O(1) for insert, remove and get random element. I draw some nice analogies with the implemention of std::vector.

Tweet card media

A Container for Insert, Delete and GetRandom in O(1) Time

I have encountered a curious problem at leetcode. You need to come up with a data structure that supports insertion, removing and retrieving a uniformly random element in average O(1) time. This...

4

4

31

@artsiom_s

Artsiom Sanakoyeu

3 years

Can Vision Transformers Learn without Natural Images? 1/ We can pretrain Vision Transformers purely on synthetic fractal data w/o any manual annotations and achieve similar performance on downstream tasks as self-supervised pretraining on ImageNet... 📝

Tweet media one

Tweet media two

Tweet media three

Tweet media four

4

8

32

@artsiom_s

Artsiom Sanakoyeu

4 years

Some nice stylization results on style transfer from our work "A Content Transformation Block For Image Style Transfer", #CVPR2019 . More result are on the project page 🌐 ▶️ 📝

Tweet media one

1

7

31

@artsiom_s

Artsiom Sanakoyeu

3 years

Germans are building a European analogue of OpenAI The German startup Aleph Alpha, which is based in Heidelberg, recently raised $ 27M . The task, they set themselves ambitious (even too much) - they want to create another breakthrough in AI, akin GPT-3.

Tweet card media

German startup Aleph Alpha raises $27M Series A round to build 'Europe's OpenAI' | TechCrunch

With Microsoft now being an investor in OpenAI the field is more open for new insurgents into the open-source AI arena. Now a German company hopes to take

1

5

31

@artsiom_s

Artsiom Sanakoyeu

4 years

PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Models #CVPR2020 Upsample photo by finding a proper latent vector in pretrained StyleGan

Tweet media one

Tweet media two

0

9

31

@artsiom_s

Artsiom Sanakoyeu

4 years

Watching @lexfridman 's podcast with @elonmusk . This is epic!

Tweet media one

1

0

30

@artsiom_s

Artsiom Sanakoyeu

5 years

Our style transfer on steroids at ICCV19! "Content and Style Disentanglement for Artistic Style Transfer" We learn subtle variations of styles and disentangle style from content Project page: Video: #ICCV19

Tweet card media

Content and Style Disentanglement for Artistic Style Transfer,...

To learn more visit https://github.com/CompVis/content-style-disentangled-STDmytro Kotovenko, Artsiom Sanakoyeu, Sabine Lang, Björn Ommer, ICCV 2019

www.youtube.com

3

16

30

@artsiom_s

Artsiom Sanakoyeu

3 years

Google open-sourced its AutoML framework for model architecture. It automatically finds the right model architecture for any classification problem. Now you can write `fit(); predict()` and call it a day! Of course, if you have enough GPUs 😅

Tweet media one

2

5

30

@artsiom_s

Artsiom Sanakoyeu

3 years

Learning High Fidelity Depths of Dressed Humans by Watching TikTok Dance Videos The single-frame depth is refined by self-supervised leveraging local transformations of body parts to enforce geometric consistency across different poses.

1

4

29

@artsiom_s

Artsiom Sanakoyeu

4 months

Check out the new paper from our team at GenAI! @AIatMeta

@AIatMeta

AI at Meta

4 months

In addition to Llama 3, today we’re also publishing a new paper: Imagine Flash: Accelerating Emu Diffusion Models with Backward Distillation ➡️ This work from GenAI researchers is enabling new image generation features in Meta AI on @WhatsApp & web.

Tweet media one

21

199

1K

1

2

29

@artsiom_s

Artsiom Sanakoyeu

3 years

ViViT: A Video Vision Transformer Pure transformer based model for video classification, drawing upon the recent success in image classification. It extracts spatiotemporal tokens from the video, which are then encoded by a series of transformer layers 📝

Tweet media one

Tweet media two

Tweet media three

0

3

29

@artsiom_s

Artsiom Sanakoyeu

3 years

Any pointers to self-supervised learning papers, where geometrical equivariance is enforced in the learned representaions? RotNet is a simple example of equivariance. "Unsupervised Part-Based Disentangling of Object Shape and Appearance" is another example. Anything else?

7

6

27

@artsiom_s

Artsiom Sanakoyeu

3 years

New Video! Computer Vision for animals is a fast-growing and very promising sub-field. I this video I explain how to reconstruct a 3D model of an animal with a single photo using a cycle consistency loss.

Tweet card media

How to Reconstruct 3D Model of an Animal from a single Photo via...

Computer Vision for Animals is one of the growing sub-fields with huge potential. In this video, I explain 2 papers for reconstructing 3D meshes of animals ...

www.youtube.com

1

8

27

@artsiom_s

Artsiom Sanakoyeu

3 years

Graph Representation Learning Book A brief but comprehensive introduction to graph representation learning, including methods for embedding graph data, graph neural networks, and deep generative models of graphs.

Tweet media one

0

8

27

@artsiom_s

Artsiom Sanakoyeu

3 years

Learning Intra-Batch Connections for Deep Metric Learning Very strong results on DML benchmarks. 🌐

Tweet media one

2

5

26

@artsiom_s

Artsiom Sanakoyeu

5 years

Happy to share that our GCPR'19 paper was selected for an oral presentation! "Semi-Supervised Segmentation of Salt Bodies in Seismic Images" : 1st place solution at TGS Salt Identification Challenge @kaggle @TGScompany Paper: #GCPR19 #kaggle

Tweet media one

0

7

26

@artsiom_s

Artsiom Sanakoyeu

3 years

Generative Adversarial Transformers 📝 🛠️ The GANsformer leverages a bipartite structure to allow long-range interactions, while evading the quadratic complexity standard transformers suffer from. Presented 2 novel attention types.

Tweet media one

1

4

26

@artsiom_s

Artsiom Sanakoyeu

10 months

Thrilled to announce that our Zurich team was directly responsible for optimizing the AI Sticker generative model. Just type in a description, and watch as it creates personalized stickers for you in IG/FB, WA. A glimpse of this in an excerpt from keynote by MZ at Meta Connect

1

2

24

@artsiom_s

Artsiom Sanakoyeu

4 years

Metric learning: cross-entropy vs pairwise losses 📝 🔨 The cross-entropy can do it better than other losses for DML. + Some theoretical explanation.

Tweet media one

0

9

25

@artsiom_s

Artsiom Sanakoyeu

3 years

MIT: Deep Learning for Art, Aesthetics, and Creativity An awesome mini-course from MIT on Neural Art and Creativity. This course has a lineup of great invited speakers like Phillip Isola (MIT), Alyosha Efros (UC Berkeley), Jeff Clune (OpenAI), etc. 🌀

Tweet media one

2

9

24

@artsiom_s

Artsiom Sanakoyeu

4 years

Unsupervised Discovery of Object Landmarks via Contrastive Learning, 2020 🖇️ The image explains the approach

Tweet media one

0

6

24

@artsiom_s

Artsiom Sanakoyeu

8 months

Check out our recent work at Meta GenAI on Accelerating the Diffusion models by caching.

@_akhaliq

AK

8 months

Cache Me if You Can: Accelerating Diffusion Models through Block Caching paper page: Diffusion models have recently revolutionized the field of image synthesis due to their ability to generate photorealistic images. However, one of the major drawbacks of

Tweet media one

1

45

220

0

0

23

@artsiom_s

Artsiom Sanakoyeu

3 years

Facebook published its ultimate SElf-supERvised (SEER) model. - They pretrained it on a 1B random, unlabeled and uncurated Instagram images 👀. - SEER outperformed SOTA self-supervised systems, reaching 84.2% top-1 accuracy on ImageNet. 🛠️

Tweet media one

2

9

23

@artsiom_s

Artsiom Sanakoyeu

3 years

MacaquePose: A Novel “In the Wild” Macaque Monkey Pose Dataset The dataset provides keypoints for macaques in naturalistic scenes, it consists of 13k images and 16k monkey instances. 📝pdf 🌀Read more in my telegram channel post

Tweet media one

1

3

23

@artsiom_s

Artsiom Sanakoyeu

4 years

> $50k to render this dataset.

Tweet media one

0

2

23

@artsiom_s

Artsiom Sanakoyeu

3 years

New video! I explain the paper "Taming Transformers for High-Res Image Synthesis". The paper introduces VQGAN which is a GAN that learns a codebook of context-rich visual parts and uses it to quantize the bottleneck representation at every forward pass

Tweet card media

VQGAN: Taming Transformers for High-Resolution Image Synthesis [Paper...

The authors introduce VQGAN which combines the efficiency of convolutional approaches with the expressivity of transformers.VQGAN is essentially a GAN that l...

www.youtube.com

2

9

22

@artsiom_s

Artsiom Sanakoyeu

5 years

New awesome image augmentation technique: CutMix #iccv2019 @naverlabseurope

Tweet media one

Tweet media two

2

10

22

@artsiom_s

Artsiom Sanakoyeu

5 years

It's such big honour to be selected as as one of the best #Neurips2019 reviewers! Moreover, I will get a free conference registration! Awesome! @hugo_larochelle

Tweet media one

2

0

22

@artsiom_s

Artsiom Sanakoyeu

3 years

Another cool work from OpenAI: Diffusion Models Beat GANs on Image Synthesis. New SOTA for image generation on ImageNet A new type of generative models is proposed - the Diffusion Probabilistic Model. 📝Paper 🛠️Code Thread 👇

Tweet media one

Tweet media two

1

11

21

@artsiom_s

Artsiom Sanakoyeu

4 years

Novel View Synthesis of Dynamic Scenes With Globally Coherent Depths From a Monocular Camera #CVPR2020 @JaeShinYoon2 This is pretty cool! The model can do space-time navigation and bullet time effect! 🔗

1

5

22

@artsiom_s

Artsiom Sanakoyeu

5 years

That's how I feel

Tweet media one

0

3

22

@artsiom_s

Artsiom Sanakoyeu

1 year

Attending #CVPR2023 tutorial Efficient Neural Networks: From Algorithm Design to Practical Mobile Deployments organized by @SergeyTulyakov @Snap

Tweet media one

Tweet media two

Tweet media three

Tweet media four

0

2

22

@artsiom_s

Artsiom Sanakoyeu

3 years

LatentCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions A framework that learns meaningful directions in GANs' latent space using unsupervised contrastive learning. 📝 🛠 Thread👇

Tweet media one

Tweet media two

Tweet media three

1

4

21