Ming-Yu Liu @liu_mingyu Twitter profile

Last Seen Profiles

@XNews

@BinorRaja

@HannahLMunday

@btwaslanisalive

@StarsSexy2

@123CardCollect

@aasandstrom

@BDesniar

@Yemen_sanaa2023

@AlfredJuntti

@niggativebts

@sanitoninn

@Anakokedamas

@soffibunda

@MontanaFranck1

@cakman62133565

@stw_pdg

@MAgentSecret

@leahfletcher3

@AusHCFJ

@bokeplokalmalam

@DianPickersgi

@KilingeD

@BinorRaja

@IAdlercreutz

@OneHuntMan

@neilminyard10

@sparklite8

@markslimdoulble

@Ta_Totoyo

@bukanashley

@OneFurAll

@melodramastuds

@BlakeBuletko

@hanapiko358

@RealAndyLeeShow

Ming-Yu Liu

@liu_mingyu

7 years

Given a content photo and a style photo, the algorithm transfers the style of the style photo to the content photo to generate a stylized output as it was captured by a camera. #DeepLearning #style Code: Paper:

76

3K

8K

Ming-Yu Liu

@liu_mingyu

5 years

The #GauGAN beta version is now available to everyone as a web service via #NVIDIA AI Playground A short illustration video is available in May everybody have fun with the app. #GAN , #SPADE

20

579

2K

Ming-Yu Liu

@liu_mingyu

5 years

A #GauGAN timelapse video created by @neilbickford from @nvidia Again, live demo is available at

15

362

1K

Ming-Yu Liu

@liu_mingyu

7 years

MUNIT: Multimodal unsupervised image-to-image translation. Learn to translate one input dog image to a distribution of cat images without paired training data. paper: code: by @xunhuang1995 @SergeBelongie @jankautz #DeepLearning

16

632

1K

Ming-Yu Liu

@liu_mingyu

6 years

Check out our #CVPR19 oral paper on a new conditional normalization layer for semantic image synthesis #SPADE and its demo app #GauGAN paper website video1 @tcwang0509 @junyanz89

13

324

872

Ming-Yu Liu

@liu_mingyu

5 years

Check out our new #GAN work on translating images to unseen domains in the test time with few example images. Live demo Project page Paper Video #NVIDIA

15

258

799

Ming-Yu Liu

@liu_mingyu

4 years

Check out our new work on face-vid2vid, a neural talking-head model for video conferencing that is 10x more bandwidth efficient than H264 arxiv project video @tcwang0509 @arunmallya #GAN

14

149

743

Ming-Yu Liu

@liu_mingyu

5 years

#GauGAN meets #3D A cool demonstration by Jay Axe, a 3D artist @nvidia Online demo is still running at

7

124

527

Ming-Yu Liu

@liu_mingyu

7 years

Snowy to summery image translation. Results from our NIPS 17 paper () Full video available in

7

207

488

Ming-Yu Liu

@liu_mingyu

6 years

code: video: Using Conditional GAN to generate HD resolution videos

GitHub - NVIDIA/vid2vid: Pytorch implementation of our method for high-resolution (e.g. 2048x1024)...

Pytorch implementation of our method for high-resolution (e.g. 2048x1024) photorealistic video-to-video translation. - NVIDIA/vid2vid

github.com

Miles Brundage

@Miles_Brundage

6 years

"Video-to-Video Synthesis," Wang et al.: Impressive stuff from NVIDIA.

3

14

75

5

136

364

Ming-Yu Liu

@liu_mingyu

5 years

In CVPR2019, the #StyleGAN paper won the best paper honorable mention and #SPADE / #GauGAN paper won the best paper finalist. Congratulates to all the GAN authors! @nvidia

8

46

342

Ming-Yu Liu

@liu_mingyu

4 years

I am looking for highly-motivated researchers that are interested and have experiences in AI for content creation (image/3D/audio/XXX) to join our research team. If interested, please send your CV to my NVIDIA email.

16

45

341

Ming-Yu Liu

@liu_mingyu

5 years

We have released our #pix2pixHD under the BSD license (It was under CC non-commercial license.) Please feel free to use it.

GitHub - NVIDIA/pix2pixHD: Synthesizing and manipulating 2048x1024 images with conditional GANs

Synthesizing and manipulating 2048x1024 images with conditional GANs - NVIDIA/pix2pixHD

github.com

1

89

325

Ming-Yu Liu

@liu_mingyu

5 years

Glad to share our #NeurIPS2019 paper on few-shot vid2vid where we address the scalability issue of our #vid2vid . Now, with 1model and as few as 1 example image provided in the test time, we could render the motion of a target subject. code coming soon.

5

99

302

Ming-Yu Liu

@liu_mingyu

6 years

Glad to see that our #GAN research works enable people to "generate realistic dance videos of NBA players for in-game entertainment." #pix2pixHD , #vid2vid

2

64

278

Ming-Yu Liu

@liu_mingyu

3 years

The #GauGAN model is now freely available in a standalone app, called #CANVAS @NVIDIAAI Visit the website and download a copy Turn your doodles into beautiful landscapes!!! #GAN

4

71

258

Ming-Yu Liu

@liu_mingyu

5 years

Woohoo! GauGAN won the Best of What's New Award by Popular Science Magazine!!! If you haven't tried GauGAN, please visit

The AI Playground from NVIDIA: Experience AI in Action

Interact with AI research demos in real-time, be inspired by the AI Art Gallery, and learn more about AI extensions in Omniverse.

www.nvidia.com

NVIDIA AI

@NVIDIAAI

5 years

The real-time #AI art sensation #GauGAN won @PopSci Magazine "Best of What's New Award" in the engineering category. See how NVIDIA researchers developed the first AI model that can produce complex images with only a few brushstrokes.

4

63

198

8

47

235

Ming-Yu Liu

@liu_mingyu

6 years

May the power of #TitanRTX with you @goodfellow_ian

2

12

227

Ming-Yu Liu

@liu_mingyu

5 years

Check out PointFlow for point cloud generation video code project Brought to you by @YangGuandao @xunhuang1995 Zekun Hao @SergeBelongie Bharath Hariharan

1

49

226

Ming-Yu Liu

@liu_mingyu

1 year

My team is hiring. If you want to join our effort in building better 3D models to digitalize the real world, here is the link for the application

Jim Fan

@DrJimFan

1 year

To give you a sense of how fast AI for 3D modeling is advancing: the field went from the left (original NeRF-reconstructed mesh) to right (Neuralangelo from NVIDIA) in 3 years. Transporting reality into high-fidelity simulation is no longer a pipe dream.

33

197

1K

2

16

223

Ming-Yu Liu

@liu_mingyu

3 years

Wondering what next after #StyleGAN -x? Check out this awesome paper from my colleagues on #AliasFreeGAN . It addresses the texture sticking issue in image synthesis and further advances the rendering quality

3

32

215

Ming-Yu Liu

@liu_mingyu

6 years

Check out our #CVPR19 oral paper on a new conditional normalization layer for semantic image synthesis #SPADE and its demo app #GauGAN paper website video1 @tcwang0509 @junyanz89

GauGAN: Changing Sketches into Photorealistic Masterpieces

A deep learning model developed by NVIDIA Research turns rough doodles into highly realistic scenes using generative adversarial networks (GANs). Dubbed GauG...

www.youtube.com

4

77

210

Ming-Yu Liu

@liu_mingyu

5 years

StyleGAN2 is out.

Analyzing and Improving the Image Quality of StyleGAN

The style-based GAN architecture (StyleGAN) yields state-of-the-art results in data-driven unconditional generative image modeling. We expose and analyze several of its characteristic artifacts,...

arxiv.org

1

41

206

Ming-Yu Liu

@liu_mingyu

6 years

We will be presenting our work on Multimodal Unsupervised Image-to-Image Translation (MUNIT) on P-1B-06 from 4-6pm on Monday at #ECCV2018 Please drop by if you want to know more details about our work. Code is available at

1

78

201

Ming-Yu Liu

@liu_mingyu

7 years

pix2pixHD code is now available in paper Brought to you by @tcwang0509 @junyanz89 @jankautz @ctnzr

3

74

187

Ming-Yu Liu

@liu_mingyu

5 years

So happy to share the news that @NVIDIADesign #GauGAN won both best real time live demo award and people’s choice of best demo award in #SIGGRAPH2019 Real-Time Live. @junyanz89 @gavriilklimov @tcwang0509 @chrisjhebert Special thank to @goodfellow_ian to bring #GAN to this world.

9

23

175

Ming-Yu Liu

@liu_mingyu

4 years

OpenAI's Dall-E is so impressive. The text2image generation capability is beyond imagination. Big congrats to @OpenAI and @ilyasut for the amazing work.

2

44

171

Ming-Yu Liu

@liu_mingyu

5 years

We will be doing a poster session for our #ICCV2019 paper on FUNIT: Few-shot Unsupervised Image-to-Image Translation () at Hall 1B. Poster 139. Hope to see you there.

4

39

165

Ming-Yu Liu

@liu_mingyu

6 years

If you want to drive a #GAN in a virtual city. Please come to #NVIDIA booth in #NeurIPS . We use #vid2vid to convert segmentation masks from a game engine to images in real-time. Steering wheel and pedal are provide.

Research at NVIDIA: The First Interactive AI Rendered Virtual World

This AI breakthrough will allow developers and artists to create new interactive 3D virtual worlds for automotive, gaming or virtual reality by training mode...

www.youtube.com

3

52

162

Ming-Yu Liu

@liu_mingyu

6 years

We made several updates to #FastPhotoStyle -20x faster now -Interface for utilizing a semantic segmentator for automatic mask generation -A new tutorial -Code -Paper will be presented in #ECCV2018 @leexiaoju #DeepLearning #style

2

47

157

Ming-Yu Liu

@liu_mingyu

5 years

I am looking for a summer PhD intern that has experience on semantic segmentation, instance segmentation, panoptic segmentation, human pose estimation, or optical flow estimation to work together on a research project. Please DM to me if interested.

4

44

150

Ming-Yu Liu

@liu_mingyu

4 years

My awesome colleagues have now released #PyTorch version of StyleGAN2-ADA. (The initial release was in #TensorFlow ) ADA uses a clever data augmentation to help address limit sample problems in #GAN training.

GitHub - NVlabs/stylegan2-ada-pytorch: StyleGAN2-ADA - Official PyTorch implementation

StyleGAN2-ADA - Official PyTorch implementation. Contribute to NVlabs/stylegan2-ada-pytorch development by creating an account on GitHub.

github.com

2

34

143

Ming-Yu Liu

@liu_mingyu

2 years

I’m looking for researchers with experiences and strong passion in large-scale image-text models to join our research team at CA. Strong knowledge on diffusion models, contrastive learning, or data curation is preferred. Team-work first, extreme hard-core, and perfection-driven.

6

23

149

Ming-Yu Liu

@liu_mingyu

3 years

We are looking for several Ph.D. interns for 2022 spring/summer/fall. We plan to cover several topics, including neural rendering, zero-shot segmentation, text--image modeling, speech/music generation, and denoising diffusion models. If interested, send your CV to me.

2

34

137

Ming-Yu Liu

@liu_mingyu

7 years

Slides for the CVPR17 GAN tutorial are now available in

GitHub - mingyuliutw/cvpr2017_gan_tutorial: CVPR'17 GAN Tutorial

CVPR'17 GAN Tutorial. Contribute to mingyuliutw/cvpr2017_gan_tutorial development by creating an account on GitHub.

github.com

0

48

129

Ming-Yu Liu

@liu_mingyu

2 years

Very happy to share #NVIDIA Picasso. It took a whole team effort to make it happen. We are hiring. DM me if you are interested in joining our effort to build SOTA foundation models.

NVIDIA Picasso, a Cloud-Based Generative AI Service for Creating...

Explore the capabilities of the Picasso image service for ISVs and enterprises to integrate simple API calls into their applications for content creation. Wa...

www.youtube.com

4

29

122

Ming-Yu Liu

@liu_mingyu

4 years

Would like to share our ECCV Spotlight paper on COCO-FUNIT: Few-Shot Unsupervised Image Translation with a Content Conditioned Style Encoder Joint work with @ksaitoboston1 and @saenko_group Project Video Code coming soon! (1/N)

5

34

120

Ming-Yu Liu

@liu_mingyu

4 years

Happy to share that our face-vid2vid together with our awesome colleagues' StyleGAN2-Ada help @NVIDIA secure the Most Disruptive Innovator award from @forbes

Forbes A.I. Awards 2020: Meet GPT-3, The Computer Program That Can Write An Op-Ed

Self-driving vehicles are now creeping into the mainstream while advances in machine learning are changing the way we write code and discover drugs. Here are the best products, the most intriguing...

www.forbes.com

6

8

117

Ming-Yu Liu

@liu_mingyu

5 years

I will be giving a talk on #GauGAN and its backbone algorithm #SPADE in room 501 at #SIGGRAPH2019 at 2pm today (Sunday 7/28). If you are interested in how we design a GAN to achieve the task and the idea behind it. Please come.

1

18

113

Ming-Yu Liu

@liu_mingyu

1 year

Very proud of the team ( @chenhsuan_lin @mli0603 Thomas Müller, Alex Evans) that brought this invention to life, which Time Magazine now recognizes as one of the best inventions of 2023. We are hiring researchers of different seniority to join our mission to democratize content

2

7

113

Ming-Yu Liu

@liu_mingyu

6 years

Going to present several #GAN works in NVIDIA’s #GTC19 conference, including #StyleGAN , #vid2vid , and several other new GAN works that we have NOT announced. Register by 2/8 for early bird pricing and use discount code NVMINGYUL for an additional 25% off:

4

24

113

Ming-Yu Liu

@liu_mingyu

7 years

MoCoGAN for mapping random vectors to videos Code/Docker Image: Paper: #DeepLearning #GAN

0

46

103

Ming-Yu Liu

@liu_mingyu

7 years

Our unsupervised image translation code is now available in Thank @goodfellow_ian for letting us using his portrait.

1

39

102

Ming-Yu Liu

@liu_mingyu

1 year

Nvidia PhD Graduate Fellowship application begins. Apply now.

0

24

101

Ming-Yu Liu

@liu_mingyu

6 years

We will present several #GAN works in NVIDIA’s #GTC19 conference, including #StyleGAN , #vid2vid , and several other new GAN works that we have NOT announced. Register by 2/8 for early bird pricing and use discount code NVMINGYUL for an additional 25% off:

NVIDIA GTC 2025 Conference Pricing

Choose the package that works best for you and your team and make sure to check out the special offers.

www.nvidia.com

0

23

99

Ming-Yu Liu

@liu_mingyu

3 years

NVIDIA Canvas Updated With New AI Model Delivering 4x Resolution and More Materials

Latest update to NVIDIA Canvas features powerful new AI model based on GauGAN2 research.

blogs.nvidia.com

1

11

88

Ming-Yu Liu

@liu_mingyu

4 years

Kudos to @tcwang0509 and @arunmallya Exciting to see a new chapter of video conferencing.

2

11

88

Ming-Yu Liu

@liu_mingyu

3 years

Congrats to @phillip_isola and @georgiagkioxari on winning the CVPR Young Researcher Awards!!!

0

3

81

Ming-Yu Liu

@liu_mingyu

7 years

Check out our MoCoGAN work on motion and content decomposed random video generation via GAN.

0

38

78

Ming-Yu Liu

@liu_mingyu

6 years

4

16

77

Ming-Yu Liu

@liu_mingyu

5 years

We are running a tutorial on deep learning for content creation in CVPR on Sunday . We have a set of amazing speakers including @phillip_isola @jtompkin Tero Kerras, @FidlerSanja Sylvain Paris @tcwang0509 @junyanz89 @elishechtman Please come to join us.

2

21

73

Ming-Yu Liu

@liu_mingyu

6 years

2

12

69

Ming-Yu Liu

@liu_mingyu

4 years

1/4 Exciting to share our #ECCV2020 paper on world-consistent #vid2vid on generating consistent renderings of 3D world. @NVIDIAAI #GAN with @arunmallya @tcwang0509 Karan Sapra paper project video

1

18

65

Ming-Yu Liu

@liu_mingyu

6 years

Congrats @tcwang0509 and @leexiaoju for their amazing works. vid2vid: FastPhotoStyle:

GitHub - NVIDIA/FastPhotoStyle: Style transfer, deep learning, feature transform

Style transfer, deep learning, feature transform. Contribute to NVIDIA/FastPhotoStyle development by creating an account on GitHub.

github.com

Yijun Li

@leexiaoju

6 years

[Vid2Vid] and [PhotoWCT] are among the 25 Best Data Science and Machine Learning GitHub Repositories from 2018~Congrats @liu_mingyu

0

10

34

2

13

65

Ming-Yu Liu

@liu_mingyu

6 years

The PWC-Net code, which won the optical flow estimation competition in the CVPR'18 Robust Vision Challenge, is now available on GitHub. Code: Paper: Challenge:

1

35

65

Ming-Yu Liu

@liu_mingyu

5 years

Want to win a #QuadroRTX 6000! Join online #GauGAN contest hosted by @NVIDIA 1. Turn your doodle into a photorealistic masterpiece with GauGAN: () 2. Share your AI artwork on Twitter with #SIGGRAPH2019 , #GauGAN & @NVIDIADesign

0

22

64

Ming-Yu Liu

@liu_mingyu

3 years

Interested in knowing more about our #CVPR2021 paper on #FaceVid2Vid , we welcome you to test out the algorithm yourself. Our online demo is available at

0

10

61

Ming-Yu Liu

@liu_mingyu

3 years

Happy to announce that GANcraft code is released!!!

Arun Mallya

@arunmallya

3 years

Code for #GANcraft (ICCV'21) has been released at , with pretrained models & training instructions. You can even import your own worlds and make them real! This also includes updates to the #Imaginaire repo to make it faster, better, and more awesome!

3

53

293

2

5

59

Ming-Yu Liu

@liu_mingyu

6 years

I will give a talk on video-to-video synthesis on #ECCV2018 Chalearn Looking at People workshop () at 5pm today. Please drop by if you are interested in knowing more details about the work. Slides available at

1

8

56

Ming-Yu Liu

@liu_mingyu

4 years

Bill is indeed a role model for many of us. While I was complaining inconvenient of working from home, he is making impacts.

0

6

52

Ming-Yu Liu

@liu_mingyu

6 years

A nice collection of papers and source code for domain adaptation research

GitHub - zhaoxin94/awesome-domain-adaptation: A collection of AWESOME things about domian adaptation

A collection of AWESOME things about domian adaptation - zhaoxin94/awesome-domain-adaptation

github.com

0

14

51

Ming-Yu Liu

@liu_mingyu

3 years

So excited to share that our deep learning-based digital avatar demo won the Best-in-Show.

ACM SIGGRAPH

@siggraph

3 years

A final congrats of the night to the jury-voted Best in Show winner, "I am AI: AI-Driven Digital Avatar Made Easy". Cheers to the team at @nvidia ! #SIGGRAPH2021 #RealTimeLive

1

16

78

5

8

49

Ming-Yu Liu

@liu_mingyu

3 years

We have made the TalkingHead-1KH dataset used in our face-vid2vid paper available. We also provide the face-vid2vid results on this dataset for benchmarking.

2

8

43

Ming-Yu Liu

@liu_mingyu

4 years

The ultimate goal of #GANcraft is to turn #Minecraft gamers into 3D artists.

Arun Mallya

@arunmallya

4 years

Introducing GANcraft, a method to convert user-created semantic 3D block worlds, like those from Minecraft, to realistic-looking worlds, without paired training data! arxiv: webpage: by @zekunhao19951 , @SergeBelongie , @liu_mingyu

11

216

793

1

4

43

Ming-Yu Liu

@liu_mingyu

4 years

@tcwang0509 and I are looking for a 2021 summer Ph.D. intern to work on an exciting image and video synthesis project. Relevant experiences on the topic and strong implementation skills are required. If interested, please DM @tcwang0509

3

11

41

Ming-Yu Liu

@liu_mingyu

4 years

Our face-vid2vid in action

Zeus Kerravala

@zkerravala

4 years

Check out video with our without Maxine. #GTC21 #GTC @nvidia

0

7

19

0

5

39

Ming-Yu Liu

@liu_mingyu

6 years

I will give a talk on multimodal image domain transfer in #ECCV2018 TaskCV workshop at 2pm (). Please come if you are interested in our recent research in the domain. Slides: Location: Room N1095ZG, Technical University of Munich

TASK-CV 2018 - Program

Accepted papers DeeSIL: Deep-Shallow Incremental Learning, Eden Belouadah, Adrian Popescu Dynamic Adaptation for Non-Stationary Visual Domains, Sindi Shkodrani, Michael Hofmann, Efstratios Gavves...

sites.google.com

0

9

40

Ming-Yu Liu

@liu_mingyu

6 years

We will be presenting #SPADE the method behind #GauGAN and #vid2vid in #GDC19 this afternoon (from 3:30pm to 4:30). Please come if you are interested. Slides are available at and

1

8

38

Ming-Yu Liu

@liu_mingyu

5 years

Since it needs as few as one example image, it can be used to make da Vinci's Mona Lisa talk.

0

11

36

Ming-Yu Liu

@liu_mingyu

1 year

Super excited about GettyImages launching their commercially safe generative AI offerings. GettyImages is a customer of NVIDIA Picasso Foundry services, and the Picasso Research team is looking for hard-core GenAI talents to join our force to build world-class GenAI capabilities

Ming-Yu Liu

@liu_mingyu

1 year

0

3

18

2

4

36

Ming-Yu Liu

@liu_mingyu

2 years

I am looking for several experienced data curation and processing engineers with our generative ai effort in the visual space. Strong knowledge of computer vision, computer graphics, and machine learning is required. If interested, please DM me.

3

7

36

Ming-Yu Liu

@liu_mingyu

4 years

We try to provide some tips on how to train your networks faster and fit in more data to your GPUs through this tutorial. We want to help you maximize your GPU investment. Feedback is very welcome.

Arun Mallya

@arunmallya

4 years

Learn how to speed up your neural network training by 2-15x with no loss in accuracy! 🤯🤩 The new and updated "Accelerating Computer Vision with Mixed Precision" tutorial is online as part of #ECCV2020 . Ft. talks by @liu_mingyu @tcwang0509 @shalinidemello

3

41

158

0

6

36

Ming-Yu Liu

@liu_mingyu

5 years

We will being giving a tutorial on how to use mixed precision (fp32 + fp16) for training deep networks for vision applications at #ICCV2019 in the Saturday morning. . Want to train your network faster with your current GPUs? Please come to check it out.

1

10

35

Ming-Yu Liu

@liu_mingyu

1 year

I will be in #CVPR2023 until end of day Wednesday. If you are interested i joining our effort, please drop me a dm. We might be able to meet in person.

Ming-Yu Liu

@liu_mingyu

1 year

My team is hiring. If you want to join our effort in building better 3D models to digitalize the real world, here is the link for the application

2

16

223

0

4

31

Ming-Yu Liu

@liu_mingyu

7 years

CVPR 2018 AI City Challenge Tracks 1> Traffic flow analysis 2> Anomaly detection 3> Multi-camera vehicle detection and re-ID

0

18

32

Ming-Yu Liu

@liu_mingyu

5 years

Glad to share the news that our #GauGAN work on translating user doodles into photorealistic images is now in exhibition in #ArsElectronica in Linz, Austria. Welcome to test it out if you are around. #NVIDIA

0

5

31

Ming-Yu Liu

@liu_mingyu

4 years

Oh yeah!!! We have 80GB memory per GPU and 640GB per node now!!! Can’t wait to use them to train larger generative models.

NVIDIA DGX Station A100 Offers Researchers AI Data-Center-in-a-Box

SC20—NVIDIA today announced the NVIDIA DGX Station™ A100 — the world’s only petascale workgroup server. The second generation of the groundbreaking AI system, DGX Station A100 accelerates demanding...

nvidianews.nvidia.com

3

1

30

Ming-Yu Liu

@liu_mingyu

4 years

NVIDIA is arguably the best place to work on the junction of computer graphics and deep learning research. Apply!

Koki Nagano

@luminohope

4 years

NVIDIA Research is hiring for Research Scientist, Deep Learning & Computer Graphics. Are you excited about generating photorealistic graphics using Deep Learning and also interested in fighting against visual misinformation? Come join @NVIDIA Research!

4

57

248

1

2

30

Ming-Yu Liu

@liu_mingyu

4 years

As many papers have their videos on youtube, it will be great if we can create a youtube video list based on the official program to simulate a more complete conference experience but free to everybody.

Ankur Handa

@ankurhandos

4 years

eccv 2020 papers are available here

1

10

44

1

3

28

Ming-Yu Liu

@liu_mingyu

3 years

Check out the recent @nvidia 's work-in-AI video #GANcraft #StyleGAN #FaceVid2Vid #DLSS #Sim2Real #SimNet #GANverse3D

NVIDIA's Work in AI

Watch to see NVIDIA's AI research in action: GANcraft Scene Generation, SyleGAN High-Res Image Generation, DLSS AI Rendering, Sim2Real Quadraped Locomotion, ...

www.youtube.com

0

5

28

Ming-Yu Liu

@liu_mingyu

6 years

AI-Based Video-to-Video Synthesis via @YouTube

NVIDIA Vid2Vid: AI-Based Video-to-Video Synthesis!

The paper "Video-to-Video Synthesis" and its source code is available here:https://tcwang0509.github.io/vid2vid/https://github.com/NVIDIA/vid2vidPick up cool...

www.youtube.com

1

10

27

Ming-Yu Liu

@liu_mingyu

7 years

Check out CASENet Full video: Paper: (CVPR'17) Code: #DeepLearning

0

3

27

Ming-Yu Liu

@liu_mingyu

4 years

Online demo available at

Ming-Yu Liu

@liu_mingyu

4 years

Check out our new work on face-vid2vid, a neural talking-head model for video conferencing that is 10x more bandwidth efficient than H264 arxiv project video @tcwang0509 @arunmallya #GAN

14

149

743

0

6

25

Ming-Yu Liu

@liu_mingyu

4 years

Check out this awesome NeurIPS paper from Tero Karras, Miika Aittala, Janne Hellsten, Samuli Laine, @jaakkolehtinen , Timo Aila on using data augmentation for training GANs with limited data. code:

NVIDIA AI Developer

@NVIDIAAIDev

4 years

NVIDIA Research developed a #GAN training breakthrough. A new technique called ADA that only needs less than a few thousand training images to generate high-resolution images. Learn how: #NeurIPS2020

1

60

228

1

24

Ming-Yu Liu

@liu_mingyu

5 years

The same model can be used for generating dancing videos of different people even for the person it didn't see during training.

2

5

23

Ming-Yu Liu

@liu_mingyu

5 years

There will be a workshop on image and video synthesis in ICCV 2019 (). Looking forward to your participation. @shiryginosar @junyanz89 @AaronHertzmann

Shiry Ginosar

@shiryginosar

5 years

Our ICCV 19 workshop - "Image and Video Synthesis: How? Why? and What If?" . Check out the excellent lineup of speakers and submit your accidental art created as a side effect of synthesis research to win NVIDIA GPUs!

0

7

33

0

7

23

Ming-Yu Liu

@liu_mingyu

3 years

Great opportunity to join my wife’s early-stage startup. She is a very dedicated, hard-working founder with great vision and execution.

0

5

22

Ming-Yu Liu

@liu_mingyu

3 years

Here is the video of our RTL demo that was selected Best-in-Show in #SIGGRAPH2021 I AM AI: Digital Avatar Made Easy SIGGRAPH Real-Time Live Demo presented by @ctnzr @arunmallya and Kevin Shih

1

5

22

Ming-Yu Liu

@liu_mingyu

4 years

Don’t miss this keynote. You will find lot of exciting new things.

NVIDIA GTC

@NVIDIAGTC

4 years

Only one hour away... Be sure to tune in for the #GTC21 keynote with #NVIDIA Founder and CEO Jensen Huang at 8:30 a.m. PDT.

8

71

241

0

22

Ming-Yu Liu

@liu_mingyu

6 years

@avsa @genekogan @tcwang0509 @junyanz89 I think it will still take sometime to reach production quality. But our engineering team will release this demo as a free online service this summer for people to try out.

2

1

21

Ming-Yu Liu

@liu_mingyu

4 years

Very proud of our colleagues. “NVIDIA won every test across all six application areas for data center and edge computing systems in the second version of MLPerf Inference.”

NVIDIA AI

@NVIDIAAI

4 years

NVIDIA extends lead on #MLPerf inference benchmark for computer vision, conversational AI, and recommender workloads, breakthrough performance enables businesses to move #AI from research to production.

0

63

194

0

1

20

Ming-Yu Liu

@liu_mingyu

4 years

Congrats to the winners of the ECCV2020 awards.

Kosta Derpanis

@CSProfKGD

4 years

#ECCV2020 awards

2

52

189

0

20

Ming-Yu Liu

@liu_mingyu

7 years

I just published “Learning Binary Residual Representations for Domain-specific Video Streaming | ”

Learning Binary Residual Representations for Domain-specific Video Streaming | mingyuliu.blog

This post briefly summarizes our work on using deep learning to improve video streaming quality for existing video compression standards…

medium.com

0

4

19

Ming-Yu Liu

@liu_mingyu

4 years

An accurate and efficient simulation environment is one key towards successful reinforcement learning research/applications. This one is exciting!

ViktorM🇺🇦

@viktor_m81

4 years

Do you want to train tasks on a single PC that was possible to solve only using big CPU clusters? Proud to share the project I've been working on for the last 2 years, end-to-end GPU accelerated simulator for RL and control:

15

74

387

0

4

18

Ming-Yu Liu

@liu_mingyu

1 year

Getty Images Launches Commercially Safe Generative AI Offering

newsroom.gettyimages.com

0

3

18

Ming-Yu Liu

@liu_mingyu

6 years

@ankurhandos @hardmaru Many training photos contain trees near water body with reflections. The #GAN eventually learns this correlation in the distribution. Interestingly, this reflection was only learned in later epochs.

1

19

Ming-Yu Liu

@liu_mingyu

3 years

Well deserved and big congrats!!!

Belongie Lab

@BelongieLab

3 years

Zekun Hao has been awarded the @NVIDIA PhD scholarship for “developing algorithms that learn from real-world visual data and apply that knowledge to help human creators build photorealistic 3D worlds. Congratulations @ZekunHao19951 @cs_cornell !

0

6

42

1

18