[1/6] Our #CVPR2025 paper “DiffusionSfM” extends our RayDiffusion framework... — inferring both geometry and cameras via diffusing pixelwise ray origins and endpoints.show more

Introducing Monitoring Beta 🔥 Seamlessly detect changes and updates... to any website Compatible with both of our scrape and crawl endpoints Join the waitlist below 👇show more

(1/2) MonoNPHM will be presented as a #CVPR2024 Highlight!... Our Neural Parametric Head Model parametrizes both geometry and appearance. With the learned model, we can then 3D reconstruct and track human heads from images or videos.show more

[1/6] What representation comes to mind when you think... of a ‘camera’? Perhaps an extrinsic + intrinsic matrix? In our ICLR (oral) paper, we instead infer a distributed representation where each pixel is associated with a ray, and show SoTA results for few-view pose estimation.show more

We are excited to launch our two models Pharia-1-LLM-7B-control... and Pharia-1-LLM-7B-control-aligned. Both models and the code used to train them are now publicly available and open-sourced for non-commercial research and educational use. Read our model blog post here: Learn more about our open-source codebase Scaling: #writtenbyalephalphashow more

Today we released the code for our CVPR 2026... paper, Flowception. Flowception bridges fully bidirectional sequence modeling and autoregressive generation by inserting frames via learned order, then denoising them with continuous flow. Website: Code:show more

This is some very cool work that builds on... our original Petri Dish NCA ( using population-based meta optimization to enrich the complexity of the system. Their paper and website are both very excellent:show more

We present “3D magician”: TADA! Text to Animatable Digital... Avatars. Given a textual description as input only, our method TADA generates expressive animatable 3D avatars with high-quality geometry and lifelike textures. (1/10)show more

Waking up Monday morning to 6-0 Michigan, 4-1 Lions,... and most of our rivals are in hell #GoBlueshow more

Rerender A Video: Zero-Shot Text-Guided Video-to-Video Translation paper page:... Large text-to-image diffusion models have exhibited impressive proficiency in generating high-quality images. However, when applying these models to video domain, ensuring temporal consistency across video frames remains a formidable challenge. This paper proposes a novel zero-shot text-guided video-to-video translation framework to adapt image models to videos. The framework includes two parts: key frame translation and full video translation. The first part uses an adapted diffusion model to generate key frames, with hierarchical cross-frame constraints applied to enforce coherence in shapes, textures and colors. The second part propagates the key frames to other frames with temporal-aware patch matching and frame blending. Our framework achieves global style and local texture temporal consistency at a low cost (without re-training or optimization). The adaptation is compatible with existing image diffusion techniques, allowing our framework to take advantage of them, such as customizing a specific subject with LoRA, and introducing extra spatial guidance with ControlNet. Extensive experimental results demonstrate the effectiveness of our proposed framework over existing methods in rendering high-quality and temporally-coherent videos.show more

Our vehicle "accident rate" declined from 33% to 14%... while our fleet grew from 200 to 300 trucks. How? We installed cameras in every truck. We implemented this change a couple years ago. As you'd expect, our drivers weren't thrilled. We tried to implement the change in a way that respected privacy while adding accountability measures that could reverse a bad trend we were seeing. For trucks assigned to a team member who drives the vehicle to and from work, we installed only road-facing cameras. Our shared trucks have dual-facing cameras that look at both the road and the cab. The cameras have sensors that can: - detect speeding, rapid acceleration, hard braking, rolling through stop signs, and distracted driving. Resistance to having a camera on board your truck is understandable, but I'm happy to say that our drivers have adapted well. And the numbers don't lie—The cameras have worked, leading to fewer accidents, fewer injuries, and less money lost to damages. Have you experienced driving a vehicle with a camera?show more

.Treasury Secretary Scott Bessent on President Trump's meeting with... Xi Jinping: "I believe that we have the framework for the two leaders to have a very productive meeting for both sides — and I think it will be fantastic for U.S. citizens, for U.S. farmers, and for our country in general."show more

And to start 2025 strong… Our work #NanoPyx is... out in Nature Methods 😍❤️ DOI: #NanoPyx is a Python framework that uses the #LiquidEngine to efficiently accelerate bioimage analysis for any user! (1/n) ⬇️show more

Excited to share our #CoRL2024 paper on humanoid motion... generation! Combining human motion priors with VLM feedback, Harmon generates natural, expressive, and text-aligned humanoid motion from freeform text descriptions. 👇(1/4)show more

Funny because if it wasn’t for a ST failure... on our end, yall woulda lost this game and gone 0-6 in the division Can’t have it both ways kiddo🤷‍♂️show more

We’ve announced clinical trial clearances from both the FDA... and Health Canada for our investigational encapsulated cell therapy for the potential treatment of type 1 diabetes. Learn more:show more

🔥 3 Homes Saved Last Night in Tustin 🔥... Just after 1:00 AM, our crews responded to a home engulfed in flames in Tustin, with fire threatening to spread to neighboring homes on both sides and to the rear. Our Incident Commander tactically deployed our firefighters >>>show more

The Power Packs Countdown has begun ⏲️ A reminder... from our co-creator Daniel Wu and more details on Seasons below💥🚀 1/6 🧵show more