🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Overleaf down? Bad timing. Or is it just me?

What happens when the NeurIPS deadline, CVPR rebuttals, and the SIGGRAPH Asia deadline all happen at the same time? Pro tip: Make sure your Overleaf syncs with dropbox or github. I'm still editing.

Amazing showing for little @Meshcapade at CVPR2025. The big companies may have more papers but I bet none of them have more papers per employee! Huge steps forward in 3D human pose and shape estimation in world coordinates, 3D hair regression from a single image, an LLM that's

Brilliant insights from Michael Black on the importance of data and 3D+ for 4D foundation models that understand humans, and the future of embodied intelligence in the last keynote talk of #Eurographics2025! See you next year in Aachen :)

Check out DiffLocks, appearing at CVPR2025. From a single image, we estimate about 100K hair strands that you can then physically simulate. We use a dataset of 40K synthetic hair images with ground truth strands. It's all available for research purposes.

This was obvious from the beginning and a big mistake by the Biden administration. I see a persistent false belief that China is a developing nation when, in reality, it is a leader in many areas of science and technology. In my area of Computer Vision, China is a powerhouse.

This lawsuit seems to have a flawed premise. The actor's union argues that the use of James Earl Jones' voice in an AI version of Darth Vader in Fortnite hurts voice actors. The fact is that this is an application that could not be done by voice actors. The AI is talking

Tomorrow is the last day of #ICRA2025, and we will be there! Don’t forget to join our Senior ML Scientist, Muhammed Kocabas, at the Vision-Based Wearable Robotics Workshop to learn all about MoCapade, a tool which captures full-body motion from any video for imitation learning

If you're an international PhD student at Harvard studying computer vision and your visa is cancelled, reach out to me or others in Europe. Don't despair. I'm sure we can find you a great place to carry on your research.

This is such a cool idea and it's amazing that it works. To train a generative talking 3D head model to have good lip sync, perform mesh to speech generation on the output. Compare this with the input speech and use the difference in your training loss. It's a neat trick and,

Next in our #CVPR2025 lineup: PromptHMR 👀✨ Drop a video and watch it blossom into crisp 3D people, even when limbs are hidden or several folks share the frame. Humans reconstructed in world coordinates with state-of-the-art accuracy 💯 By Yufu Wang, Yu Sun, Priyanka Patel,

Estimating 3D humans from video is typically seen as "pixels to parameters" regression problem. Here we add side information to prompt the process. These prompts can come from a user or an LLM. Prompts are things like bounding boxes around the body, face detections, segmentation

My dream has been to take a photo of a person and extract the 2D sewing pattern of their clothing and then turn it into a 3D garment. ChatGarment does exactly this plus it lets you edit the garment, or create a completely new one, using text prompts. chatgarment.github.io At

Next in our #CVPR2025 series: ChatHuman 🧠🕺 By Jing Lin, Yao Feng, Weiyang Liu, and Michael J. Black, in collaboration with the Max Planck Institute for Intelligent Systems (Intelligent Systems), Stanford University (Stanford University), Tsinghua University (Tsinghua University), and University of

Capturing 3D human pose and motion has made great progress but humans interact with 3D objects. Training robots from demonstration requires that we estimate the human, the object, and the contacts in 3D. PICO is a step towards that goal. At CVPR2025.

🚀 Headed to #CVPR2025? Want to create 3D garments faster and at massive scale? This is your chance to see how synthetic data is reshaping the game. Join our brilliant CG Scientist, Maria Korosteleva, as she takes the stage at the CVPR Workshop on Virtual Try-On 💃 🧵 Breaking