Datagen is powering the AI revolution by providing high-performance synthetic data, with a focus on data for human-centric computer vision applications.
New York
Joined June 2021
- Tweets280
- Following96
- Followers411
- Likes59
Datagen retweeted
Unbox the world of AI!📦
@GlennJocher joined @RealGilbaz on @DatagenTech's Unboxing AI podcast where he discussed his journey into CV and YOLOv8 tech specs while offering insights on other popular open-source models like ChatGPT, CLIP, and LLaMA.
Tune in wherever you listen to podcasts: bit.ly/3MPov26
Datagen retweeted
Upgrade your expressive 3D human avatars from #SMPL-X to #SUPR, our latest and greatest body model. SUPR is trained from 1.2M 3D scans, is more expressive, and includes feet with articulation and compression. Code by @NeelayShah8, video by @AYiannakidis. github.com/NeelayS/supr_conv…
Datagen retweeted
Incase you missed it, here's the recording of our latest webinar on The Future of Generative AI and Synthetic Data with @RealGilbaz from @DatagenTech.
youtube.com/watch?v=uJ9ShAG8…
Datagen retweeted
We have Ofir Zuk on the Infinite ML pod today. He's the cofounder and CEO of Datagen.
We talk about:
- the role of synthetic data in AI
- methods used to generate synthetic data
- Generative Adversarial Networks
- synthetic vs simulated data
- fidelity vs privacy
Datagen retweeted
💥@_segmind is hosting The Future Of Generative AI and Synthetic Data: @DatagenTech and Segmind
🤩Get ready to explore the future of AI and Synthetic Data.
⚡️Would you like to attend?
✅Sign up with Linkedin for Free.
linkedin.com/video/event/urn…
#generativeAI, #syntheticdata, and #segmindfuture. 🚀Let’s go!
Datagen retweeted
Pic2Word: Mapping Pictures to Words for Zero-shot Composed Image Retrieval
abs: arxiv.org/abs/2302.03084
github: github.com/google-research/c…
Datagen retweeted
StyleGAN-T: Unlocking the Power of GANs for Fast Large-Scale Text-to-Image Synthesis
significantly improves over previous GANs and outperforms distilled diffusion models in terms of sample quality and speed
abs: arxiv.org/abs/2301.09515
project page: sites.google.com/view/styleg…
Datagen retweeted
Chatting with @geoffreyhinton was really inspiring, so excited to get to share our conversation:
Apple: apple.co/3NLtQED
Spotify: spoti.fi/3GBDpDM
Google: bit.ly/3aD7ZkN
Youtube: youtu.be/2EDP4v-9TUA
Datagen retweeted
Vall-E AI voice tool can mimic human speech in seconds #VALL-E shacknews.com/article/133631…
Datagen retweeted
My top paper tweets in 2022 🧵
A ConvNet for the 2020s
abs: arxiv.org/abs/2201.03545
github: github.com/facebookresearch/…
@huggingface demo: huggingface.co/spaces/akhali…
Datagen retweeted
Data-centric #computervision gains more traction in 2022, with exciting work by AI/ML companies such as @gretel_ai, @DatagenTech, @tonicfakedata, @wandb, @CleanlabAI, @labelbox, @LabelStudioHQ, #CVAT.AI, @scale_AI & @V7Labs. Get the '22 CV recap:
medium.com/voxel51/why-2022-…
Datagen retweeted
Excited to see both DALLE (Diffusion) & Instant Rendering (NeRF) here - shows how Machine Le)arning is becoming more and more mainstream 😀
time.com/best-inventions-202…
Datagen retweeted
The Sintel story: build what you need & use what you build. If you need it, others probably do too. If you don't use it, don't expect anyone else to. My new blog post tells the backstory of the Sintel dataset and why I think it has stood the test-of-time. perceiving-systems.blog/en/n…
Datagen retweeted
UniTune: Text-Driven Image Editing by Fine Tuning an Image Generation Model on a Single Image
abs: arxiv.org/abs/2210.09477
Datagen retweeted
Top Synthetic Data Tools/Startups For Machine Learning Models
marktechpost.com/2022/10/08/…
#ArtificialIntelligence #DataScience @DatagenTech @parallel_domain @SynthesisAI_ @YData_ai @mostly_ai @OneView_AI @StaticePrivacy @skyengineai @hazy_ai @tonicfakedata @gretel_ai
Datagen retweeted
SUPER excited to present ShAPO🎩, our #ECCV2022 work on real2sim asset creation and category-level 3D object understanding! Done during my internship at @ToyotaResearch!
Webpage: zubair-irshad.github.io/proj…
@GTrobotics @ICatGT @mlatgt @eccvconf
⬇️(1/9)
Datagen retweeted
This piece from @DatagenTech is a great intro on facial landmarks used to identify and represent key parts of a human face - widely used for computer vision applications.
hubs.ly/Q01gclnc0
Datagen retweeted
Can fake data actually do good? More tech companies are using 'synthetic' images to train their AI to be more fair and accurate. Take this simulation of a woman falling asleep at the wheel, created by @DatagenTech. It's one of thousands being used to train in-car sleep sensors.
Datagen retweeted
Our code for vision-aided GANs has been released (github.com/nupurkmr9/vision-…). It significantly improves StyleGAN2/3 and BigGAN. We can aid generative modeling by leveraging the knowledge from large-scale pre-trained recognition models (e.g., CLIP, DINO, Swin-T).
Short on GPUs/time/data/$, but still want to train a GAN?
In "Ensembling Off-the-shelf Models for GAN Training", we show GANs can be "Vision-Aided" with pre-trained nets as discriminators.
A #CVPR2022 (oral) by @nupurkmr9, @elishechtman, @junyanz89
Web: cs.cmu.edu/~vision-aided-gan…