Ever wondered what it's like to work at Runway? Check out today's blog post for a peek into the interview process and day to day of our interns Sharon Lee and Peter Michael: bit.ly/3iTL7xW And yes — we're hiring!

So true 😭

Very exciting 'breaking' news! CompVis (research group behind VQGAN) have just released a new 1.45B parameter model to its Latent Diffusion model: github.com/CompVis/latent… From the released image it seems like it has an unprecedented text-synthesis capacity. More to follow soon

Collaborative video editing is here! 👨‍👨‍👦‍👦 Real-time multiplayer editing 🔗 Invite collaborators with a single link 🌍 Make projects public 🏢 Create Teams (early beta coming soon!) Start collaborating now: runwayml.com

Automate the tedious. Focus on creativity. Incredible Runway research collab with between Patrick Esser Roni Sengupta Peter Michael Allen School

AI can unlock joint human/computer creativity! Imagen is one direction we are pursuing: gweb-research-imagen.appspot.com "A high contrast portrait of a very happy fuzzy panda dressed as a chef in a high end kitchen making dough. There is a painting of flowers on the wall behind him."

Feel pleased to show off our upcoming #CVPR2025 paper, HumanNeRF. Our method turns a Youtube video of moving people into a free-viewpoint video! Page: grail.cs.washington.edu/projects/human… Video: youtu.be/GM-RoZEymmw with Brian Curless, Pratul Srinivasan, Jon Barron, ira kemelmacher More in thread! (1/7)

We present Latent Diffusion Models in tomorrow's Image & Video Synthesis and Generation session #CVPR2025 in Hall B1 (Thur. 08:30-10:18). Join us for a chat at Poster #7 between 10:00-12:30 in Hall B2-C and sample some jazz! With Robin Rombach Andreas Blattmann D. Lorenz Runway B. Ommer

📢📢📢 Thrilled to introduce "𝐌𝐨𝐛𝐢𝐥𝐞𝐍𝐞𝐑𝐅: exploiting the polygon rasterization pipeline for efficient neural field rendering on mobile architectures" → with 𝐥𝐢𝐯𝐞 𝐝𝐞𝐦𝐨𝐬 arxiv.org/abs/2208.00277 (Internship project lead by Zhiqin Chen)

I'll tell you a secret about having more impact. You've got to focus on fewer things, ideally just one -- your most ambitious idea. And you've got to commit and finish it.

Is there a doctor on the plane? -Yes, but not that kind of doctor. -The passenger in 36c is trying to inpaint an image using a pre-trained stable diffusion model and simply copy-pastes the inpainting observed part in place, after each iteration! -Ok, I got this.

wild.

At today's Image Analysis section @DIKU @UCPH workshop, section head Kenny Erleben asked me to speak about "how to get cool research ideas." (1/25)

An app developed by researchers in Abe Davis’ group uses augmented reality to help users repeatedly capture images with a phone or tablet from the same location to make time-lapse videos. ReCapture is now available in the Apple app store. Read more: bit.ly/3tDxHMa

Ray Conditioning: Trading Photo-consistency for Photo-realism in Multi-view Image Generation abs: arxiv.org/abs/2304.13681 project page: ray-cond.github.io

Noah Snavely Noah Snavely, associate professor of computer science at @Cornell_Tech & Cornell Computer Science, and a team of researchers have developed OmniMotion, a new video optimization tool that estimates motion across a full video with only a small sample of frames. news.cornell.edu/stories/2023/1…

If you have a career in research, or if you aspire to have a career in research, you must watch (or read) Dick Hamming's "You and Your Research". Easily the best advice out there on the subject. Video: youtube.com/watch?v=a1zDuO… Transcript: cs.virginia.edu/~robins/YouAnd…

Detecting a fake: A team of Cornell Bowers Computing and Information Science researchers has developed a way to “watermark” light in videos to detect if video is fake or has been manipulated, another potential tool in the fight against #misinformation. Abe Davis Serge Belongie ACM SIGGRAPH news.cornell.edu/stories/2025/0…

This was such a fun project to work on! Check it out!

Come to our #SIGGRAPH2025 talk today at 2pm at the Computational Imaging, Lighting & Display session in the West Building rooms 118-120 to learn more!