Today, I’m thrilled to share the public launch of AXL, a new venture studio building the next generation of AI-powered companies, right here in Canada.

Congrats to RI Ph.D. Benjamin Attal, RI assistant professor Matthew O'Toole & co-authors from University of Toronto & Vector Institute for winning the CVPR 2025 Best Student Paper award! 🙌 Check out the winning work: "Neural Inverse Rendering from Propagating Light" 💡🏆 loom.ly/2dS1PJw

Starting a new research project? Don’t let fuzzy goals derail you. Check out the Heilmeier Catechism as one way to clarify and assess your research plan. Thanks David Lindell for sharing!

📸Introducing "Generative Panoramic Image Stitching"📸 Traditional panorama stitching struggles with casually captured images. We fix that by fine-tuning diffusion models to in/outpaint and stitch for us. website: mathieutuli.com/generative-pan…… arxiv: arxiv.org/abs/2507.07133

📸 Join us at ICCV 2025 for our workshop on Computer Vision with Single-Photon Cameras (CVSPC)! 🗓️  Sunday, Oct 19, AM – PM at the Hawai'i Convention Center. 🔗 Website: cvspc.cs.pdx.edu. 🗣️ Invited Speakers: Mohit Gupta, Matthew O'Toole, Dongyu Du, David Lindell,

#ICCP2025 Paper Awards 🏆

👏 HUGE congrats to the #ICCP2025 award winners for their outstanding contributions!

Highlights from our 2-day #ICCP2025 (hands-on) summer school.

Kicking off #ICCP2025! Highlights from Day 1 of the oral sessions.

Exploring cutting-edge computational imaging research at the #ICCP2025 U of T Department of Computer Science lab tours 🥼

Happy to have led local arrangements for #ICCP2025! Massive thanks to our volunteers for making it possible: Dongyu Du, andrew guo, Len Luong, Ali SaraerToosi, Howard Xiao, Andrew Xie, Sophia Yang, and Kelly Zhu! And, Sotiris Nousias and Jazmin Diaz for being the best co-chairs!

I recently received the Ontario Early Researcher Award — thank you to the Province of Ontario for your support!

🚨 Code Release for CAP4D🧢 Excited to share that we have released the inference code and weights for CAP4D! Visit github.com/felixtaubner/c… to generate an avatar and view it just like this one (yup that's me)

Checkout this episode of the Executive Code Podcast for a dive into our work on modeling light propagation!

Excited about this work led by Esther Lin — we calibrate and visualize high-dimensional point spread functions (PSFs) for a range of cameras and lenses. A surprising result is that PSFs from two different iphone cameras of the same model have subtle and detectable variations!

This is Ray3. The world’s first reasoning video model, and the first to generate studio-grade HDR. Now with an all-new Draft Mode for rapid iteration in creative workflows, and state of the art physics and consistency. Available now for free in Dream Machine.

Looking forward to this workshop at #ICCV2025!

Really exciting results on 3D and 4D generation without using any multiview datasets---only supervision from a video model! Collaboration with the amazing folks at NVIDIA, Andrea Tagliasacchi 🇨🇦, and led by Sherwin Bahmani

Check out MVP4D — single image to controllable 360-degree avatars using multi-view video diffusion models! Led by Felix Taubner felixtaubner.github.io/mvp4d/

Very interesting questions posed by our last three speakers: “How far can we go with synthetic data?” “What can we learn from propagating light?” “How ambiguous is lighting in images?” Thanks to the speakers: Julien Philip , David Lindell and Ko Nishino! #ICCV2025