Feature 3DGS adopts the same 3D Gaussian initialization from sparse SfM point clouds as utilized in 3DGS, with the addition of an essential attribute: the semantic feature. Feature 3DGS enables the distillation of 3D feature fields from any 2D foundation models, using only 18

Our paper on adding explicit constraints to diffusion models will be at the “Magic Diffusion” session of Siggraph Asia. Enables geometrically accurate images, and also some unexplained semantic changes. Led by UCLA PhD student Rishi Upadhyay

Professor Kyros Kutulakos has received a 2023 Dean’s Research Excellence Award from U of T Arts & Science. His computer vision and computational photography research is widely admired for its creativity, rigour and transformational impact. 👏 uoft.me/9-7

fun story: terry tao was on both my and my brother's committee. he solved both our dissertation problems before we were done talking, each of us got "wouldn't it have been easier to...outline of entire proof" 🫠

+1 to Ben P and between bill freeman’s famous (and accurate) plot and never-ending complaints about the randomness of the review process… can we really blame people for coming to the misguided conclusion “draw as many samples as I can to maximize chance of getting an outlier”?

Our team at Google is hiring research interns for 2024! If you are interested in 3D generation / reconstruction, please apply!

UCLA is poised to take over property in West Los Angeles once primed for offices of Google. bizjournals.com/losangeles/new…

Hello #ces2024! We are excited for a week of meetings and demos with media and customers! #robotics #ai #autonomousmobility

Foundation models raise questions on "forward compatibility" of sensors. A disruptive, new sensor is not well represented in training data for FM, whereas legacy sensors are more compatible Pradyumna Chari seminar topic in UCLA-Caltech Memorial Lecture tmw at 12 noon PT.

🤖 Attending #MODEX2024? Don't miss hearing from Nitish Srivastava this Wed on a panel with NVIDIA AI discussing #ai and #lastmiledelivery. See you there! For more info on the session: lnkd.in/gm-nt84c #robotics #ai #lastmiledelivery #materialhandling

Ever wondered how humans would perceive the world with eyes 1000x faster, sensitive to infrared, or capable of "seeing" coffee cup vibrations from our voice? For humans, it's a mystery. But for 𝗺𝗮𝗰𝗵𝗶𝗻𝗲𝘀, my group at WIS aims to answer these ques., and you can take part!

Our fantastic teacher Petros Boufounos (Petros Boufounos) in action during the 6.347 class (MIT EECS, circa 2014). What an amazing class + cohort! ... yet to find a book that matches the breadth and depth of Signal Processing as it was taught here. I hope 6.347 becomes a book one day!

DreamScene360 Unconstrained Text-to-3D Scene Generation with Panoramic Gaussian Splatting The increasing demand for virtual reality applications has highlighted the significance of crafting immersive 3D assets. We present a text-to-3D 360^{circ} scene generation pipeline

Very interesting spotlight paper at cvpr that can open up more research in exploiting more advanced eyes that animals have. Congratulations to the POSTECH and Princeton team

I'm thrilled to share that I will be joining Purdue University as an Assistant Professor in the Department of Computer Science in Fall 2025!!! Many thanks to all my mentors, colleagues, and friends for their support. I will be hiring students this upcoming cycle. Please get in