Can AI make graphics look realistic ? 🚗 👇Proud to announce a new work with Stephan Richter and Vladlen Koltun. I worked on this in my internship at Intel but majority of credit is to Stephan who persisted to make it work and work robustly. full video: youtu.be/P1IcaBn3ej0

🤖 Hiring PhD/Master Research Intern (m/f/d)🤖 If you are (or know someone) interested in doing state-of-the-art 3D-GAN/Neural Rendering Research that will change REAL design applications, DM me or write us at: talent@pch-innovations.com Bonus: Fresh snacks & V100 GPUs 😉

PhD student and looking for an #internship? Sink your teeth into juicy #MachineLearning and #ComputerVision problems right at the frontier of #AugmentedReality at @NianticLabs! careers.nianticlabs.com/open… Spread the word! ❤️❤️❤️

Great #GCPR2020 talk from Felix Draxler about the role of single affine coupling blocks in normalising flows. The group in Heidelberg has certainly nailed the accessibility of their talks and explanations, I'm a big fan! Certainly worth your 10 minutes. youtube.com/watch?v=_T6I0PSg…

Major update and code release for our #relocalization pipeline, DSAC* (Updated) paper: arxiv.org/pdf/2002.12324.pdf Code: github.com/vislearn/dsacstar Use RGB or RGB-D, train with an SfM model or a 3D scan. Train from images and poses alone and DSAC* discovers coarse geometry by itself:

We will be doing a live demo and Q&A of @blankcompose at @eccvconf today at 15:00 CET. Join us at #ECCV2020 virtual platform at [Networking lounge => demonstrations => Blank Compose] #GAN #BlankCompose #AI

Interested in learning about normalizing flows in computer vision? Check out our tutorial at ECCV 2020: mbrubake.github.io/eccv2020-… Videos are live now for registered attendees and live sessions are Sunday at 5:30am and 4pm ET. Drop in and ask us anything! @Carsten_Rother

#AI can now create high-quality 3D car design🚗, product design👟, or game character🧝‍♀️ in 20 milliseconds ! 🤖 I'm proud to share our work with #pchinnovations on @blankcompose, a state-of-the-art controllable 3D generative model. Detailed video in next tweet👇. DM for live demo

Checkout our work on "Self-Supervised Viewpoint Learning from Image Collections" at #CVPR2020 Showing from 4:00 - 6:00 AM (PST). research.nvidia.com/publicat…

What to do if your re-localization method works for small environments but not big ones? Cut it into small pieces of course! Straight forward, but interesting implications if you still want to train everything jointly and end-to-end. #ICCV2019 #ICCV19 #ICCV #DeepLearning

We released the @PyTorch version of "Deep Object Co-Segmentation" too. you can find it at github.com/ohosseini/DOCS-py….

Full house for the great talk by @JeffDean at the Heidelberg.ai meeting.

We've released the #Caffe version of our work "Deep Object Co-Segmentation". It's a joint work with Omid Hosseini Jafari (@o_hosseini ) at @LabHeidelberg. Source-code: github.com/ohosseini/DOCS-ca… … Project-page: ohosseini.github.io/projects… We will release the @PyTorch version soon.

I'm excited to be visiting the University of Heidelberg to give a talk this Tuesday, 25 September, 2018 at 1pm.

Deep Object Co-Segmentation arxiv.org/abs/1804.06423 #computervision

New cool work from our group about Invertible Neural Networks (INNs)