SIGGRAPH Asia 2026

Headlines · 2

1. Oct 20
   RealMe uses video diffusion to make 2K, 4D-consistent volumetric video from sparse captures

   A paper titled "RealMe: Immersive Volumetric Video from Minimal Captures," by Chenxu Zhou and co-authors including Xiaowei Zhou, was submitted to arXiv on 21 October 2026 and is slated to appear in SIGGRAPH Asia 2026 Conference Papers. The authors describe RealMe as a unified framework built around a high-resolution video diffusion model that enhances sparse, low-resolution captures into 2K, 4D-consistent video for volumetric reconstruction, using auxiliary data encoded into latents with layout-aware positional embeddings and a latent frequency loss. They report experiments on the public DNA-Rendering benchmark and their own RealVV dataset showing better visual fidelity and 4D consistency than prior methods, though these results are author-reported and not independently replicated.

   Hugging Face · Papers · 🔥 0
2. Oct 20
   UniMate open-sourced: one model animates diverse skeletons

   UniMate, a research project from Princeton, UC Berkeley, MIT and NTU, has released training and inference code under the MIT license plus preview checkpoints under CC BY-NC 4.0 for a unified text-to-animation model accepted to SIGGRAPH Asia 2026. It also released UniML3D, described as 13,769 text-paired motion clips across 7,430 skeletons spanning bipedal, quadrupedal, avian, marine, insectoid and serpentine rigs as well as articulated rigid objects. The README says inference generates motion for arbitrary rigged assets in real time without per-skeleton retraining or test-time optimization, and an October 5 rig_preprocess tool converts GLB, glTF or FBX assets; the authors acknowledge that many motions and skeletons still fail.

   GitHub Trending · Python · 🔥 0

Experience and discussion from the community

Share my SIGGRAPH Asia 2026 experienceAsk about SIGGRAPH Asia 2026