#ECCV2024 We have released the evaluation script, pretrained models, and rendered results of Generalizable Human Gaussians (GHG). Code: github.com/humansensingla… Paper: arxiv.org/abs/2407.12777 Website: humansensinglab.github.io/Generalizable-…

📢Announcing our 3D head avatar benchmark📢 Two tasks with hidden test sets: - Dynamic Novel View Synthesis on Heads - Monocular FLAME-driven Head Avatar Reconstruction Our goal is to make research on 3D head avatars more comparable and ultimately increase the realism of

📢 We just released the code for JetFormer at github.com/google-researc… Enjoy!

📣📣📣 Excited to share GENMO: A Generalist Model for Human Motion. Words can’t perfectly describe human motion—so we build GENMO. It’s everything to motion. 🔥Video, Text, Music, Audio, Keyframes, Spatial Control…🔥 -- GENMO handles it all within a single model. 📹 Two

Fine-tune 100+ LLMs directly from a UI! LLaMA-Factory lets you train and fine-tune open-source LLMs and VLMs without writing any code. Supports 100+ models, multimodal fine-tuning, PPO, DPO, experiment tracking, and much more! 100% open-source with 50k stars!

🚨 Code dropped! One of the most exciting papers of the year! Check it out! Link in the comments 👇

Discover the right 3D Geometric Foundation Model for your task—whether it’s stereo matching, multi-view depth estimation, video depth, pose estimation, semantic understanding, or novel view synthesis. Explore more insights in our #E3DBench #FoundationModel #3D #GaussianSplatting.

ChatHuman: Chatting about 3D Humans with Tools TLDR: A method for integrating many tools for digital human analysis with an LLM. This method uses RAG on academic publications to help an agent select appropriate tools based on context. 📽️ Project Page: chathuman.github.io 📜

Huge thanks to the DeepBeepMeep team deepbeepmeep for their support! Now, you only need 10 GB of VRAM—instead of 80 GB or 24 GB—to generate 15-second voice/song-driven videos. It’s fast, with no quality loss, and supports TeaCache. 🤗Hugging Face: huggingface.co/tencent/Hunyua…

Join us at the workshop on Photo-realistic 3D Head Avatars at #CVPR2025, featuring invited talks by leading academic researchers and industry professionals at the forefront of avatar technology. Wednesday 11th, starting from 1:20pm workshop website: kaldir.vc.in.tum.de/nersemble_benc…

Congrats to the #VGGT work for winning the best paper award at #CVPR2025! Very proud to have David Novotny our CTO SpAItial AI among the contributors -- congratulations!

Upcoming CVF-related #computervision conferences.

The recordings from the #CVPR2025 tutorial "Volumetric Video in the Real World" are now available! nvlabs.github.io/cvpr2025-volum…

The materials of my graduate school course at #SGP2025 on "Deep Learning on Meshes and Point Clouds" are live on rubenwiersma.nl/deeplearning! A 'map' for 3D deep learning, covering the basics and thoughts on when and where to use deep learning tools in 3D.

Recent update of🚀awesome-dust3r: github.com/ruili3/awesome…🚀 - A new trend of 3D geometric reasoning: LaRI (ruili3.github.io/lari), RaySt3R (rayst3r.github.io), Amodal3R (sm0kywu.github.io/Amodal3R/). - A new direction for DUSt3R4Science: CryoFastAR (arxiv: 2506.05864).

For everyone interested in precise 📷camera control 📷 in transformers [e.g., video / world model etc] Stop settling for Plücker raymaps -- use camera-aware relative PE in your attention layers, like RoPE (for LLMs) but for cameras! Paper & code: liruilong.cn/prope/

Your 3D Gaussian Splats look great, but physics? They've been a flop. Discover how to do deformable simulations on 3DGS at #NVIDIAResearch's NVIDIA Kaolin & Warp workshop at #SIGGRAPH2025 👉 nvda.ws/4o61SGP

Reconstructing 4D Spatial Intelligence: A Survey Yukang Cao, Jiahao Lu, Zhisheng Huang, Zhuowei Shen, Chengfeng Zhao, Fangzhou Hong, Zhaoxi Chen, Xin Li, Wenping Wang, Yuan Liu, Ziwei Liu tl;dr: in title arxiv.org/abs/2507.21045

"Cameras as Relative Positional Encoding" TLDR: comparison for conditioning transformers on cameras: token-level raymap, attention-level relative pose encodings, a (new) relative encoding Projective Positional Encoding -> camera frustums, (int|ext)insics for relative pos encoding

Another major update of the "awesome-dust3r" (github.com/ruili3/awesome…) paper list. There are more VGG-T follow-ups and some interesting correlations, e.g., VGGT-Long/LONG3R, STream3R/Streaming 4D-VGGT. Let's see what happens for visual geometry in the DINO-v3 era :)