Guoxing Sun retweeted
🚀 Meet Orbit: OFT-based RL infrastructure for stable, efficient post-training of trillion-parameter LLMs.
Orbit can train 1T+ LLMs (e.g. Kimi-2.6, DeepSeek-V4-Pro) on a single GPU node (8xB200) with extremely small train-rollout gap!
Code: github.com/Sphere-AI-Lab/orb…
Blog: spherelab.ai/orbit/
Blog in Chinese: mp.weixin.qq.com/s/M3Q4AnhMa…
Guoxing Sun retweeted
I’m happy and honored to share that I’ve received the Eurographics Young Researcher Award 2026.
My sincere thanks to Eurographics and the award committee for this recognition.
Guoxing Sun retweeted
Excited to share our CVPR’26 work! Moving beyond dense captures to long-tail Internet photos, we introduce MegaDepth-X and sparsity-aware sampling for 3D reconstruction.
Great work led by Yuan (@yuanli16342871) and the team.
Project page: megadepth-x.github.io
Introducing: Long-Tail Internet Photo Reconstruction (CVPR’26)
We go beyond densely captured imagery to train more general 3D foundation models for the long tail of noisy, sparse, incomplete Internet photo collections of 3D scenes. Yet, we face a data bottleneck: models need ground truth for these long-tail scenes, which classical SfM fails to provide. How do we bypass it?
We break this bottleneck with two key contributions. First, we introduce MegaDepth-X, a large new dataset of scenes with high-quality 3D supervision. Second, we propose a new way to simulate difficult image sets for training.
Project page: megadepth-x.github.io/
Guoxing Sun retweeted
If you are at #Eurographics tomorrow, don't miss our STAR session on "How to Build Digital Humans?" 🕺
🗓️ Monday, 4th of May
🕐 1:15 pm - 2:45pm
🏡 Kino 5
We will have experts in the field share their thoughts on 3D avatars.
It will be cinematic!
Guoxing Sun retweeted
D-Rex by NVIDIA.
photorealistic digital humans that move and look correct under any lighting condition.
- built on EVA +Cosmos DiffusionRenderer
- supports facial expression,3D pose control under novel HDR lighting.
so, Hollywood-grade relightable characters without the usual technical headaches of traditional 3D rendering pipelines.
vcai.mpi-inf.mpg.de/projects…
Guoxing Sun retweeted
🚨 Do 3D-LLMs really understand 3D? Short answer: no.
Excited to share Real-3DQA at #ICLR2026!
- Project: real-3dqa.github.io
- Arxiv: arxiv.org/abs/2603.23523
- Data: huggingface.co/datasets/Oliv…
🎥 1-min video explainer ↓
Guoxing Sun retweeted
Thrilled to announce that our workshop proposal, "See the World🌍 in a Different Light ✨", has been accepted for #CVPR2026 in Denver! 🎉🎉🎉
Guoxing Sun retweeted
📢Seen2Scene
Real-world 3D is incomplete, typically requiring training on synthetic scene data.
@QTDSMQ introduces visibility-guided flow matching, enabling training on real partial scans for scan completion & text-to-3D scene generation!
Check it out: quan-meng.github.io/projects…
Guoxing Sun retweeted
Figure 03. Impressive. 🤖🦾🦿@adcock_brett
Guoxing Sun retweeted
Photos to 3D model with 2Xplat.
Pose-free 3D Gaussian Splatting
- 0.15s–0.31s inference;
- beats YoNoSplat;
- high-fidelity splats
hwasikjeong.github.io/2Xplat…
Guoxing Sun retweeted
NeuralFur wins Best Paper Runner Up at @3DVconf. From multi-view images, we create a strand-based hair groom for animals. Unlike human hair, fur varies in length across the body parts of animals.
NeuralFur leverages a VQA approach to infer fur lengths and directions across the body and to create a furless mesh.
We then reconstruct strand-based fur geometry from multi-view images, resulting in a realistic animal model that is ready for physics-based animation in game engines like Unreal.
Code is online. Check out the project page link below.
Congratulations to @ness_pirs @bernakabadayi @AYiannakidis @gfgbec and @JustusThies!
neuralfur.is.tue.mpg.de/
Guoxing Sun retweeted
Whole body motion tracking open source code: github.com/Axellwppr/motion_…
Guoxing Sun retweeted
🤖Simulation-Ready Human–Scene Interaction Reconstruction🤖
#HSImul3R reconstructs sim-ready 3D human-scene interactions from casual videos with *physics-in-the-loop* that can be directly deployed to humanoids.
- Project: yukangcao.github.io/HSImul3R…
- Code: github.com/yukangcao/HSImul3…
Guoxing Sun retweeted
We have seen many works unlock the power of pretrained models for images and videos🏞️. But what about human motion🕺💃?
Can we leverage a pretrained motion prior for a wide range of downstream tasks?
Yes!! UMO is a simple yet effective framework that, for the first time, unlocks the priors of a motion foundation model (i.e., HY-Motion) for 10+ tasks, including editing, reaction generation, stylization, trajectory control, obstacle avoidance, keyframe infilling, and more. Amazing work! @xiaoyan_cong and @kunkun0w0.
🏠Webpage: oliver-cong02.github.io/UMO.…
📄 Paper: arxiv.org/abs/2603.15975
With the growing number of tools for transferring SMPL motion to humanoids, we hope it could also become a source of skills for humanoid robot learning.
#Graphics #Motion #Animation #AIGC #GenerativeAI #Vision #3DV #Robotics #Robot #Humanoid #Learning #GenAI #Animation
💡Introducing 𝑼𝑴𝑶 -- one unified model that unlocks motion foundation model (HY-Motion @TencentHunyuan) priors for 𝟏𝟎+ 𝐭𝐚𝐬𝐤𝐬: 𝐞𝐝𝐢𝐭𝐢𝐧𝐠, 𝐫𝐞𝐚𝐜𝐭𝐢𝐨𝐧 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧, 𝐬𝐭𝐲𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧, 𝐭𝐫𝐚𝐣𝐞𝐜𝐭𝐨𝐫𝐲 𝐜𝐨𝐧𝐭𝐫𝐨𝐥, 𝐨𝐛𝐬𝐭𝐚𝐜𝐥𝐞 𝐚𝐯𝐨𝐢𝐝𝐚𝐧𝐜𝐞, 𝐤𝐞𝐲𝐟𝐫𝐚𝐦𝐞 𝐢𝐧𝐟𝐢𝐥𝐥𝐢𝐧𝐠... (1/8)
🌐 Webpage: oliver-cong02.github.io/UMO.…
📄 Paper: arxiv.org/abs/2603.15975
Guoxing Sun retweeted
(1/5)
🚀 Excited to share our CVPR paper MATCH!
Website: malteprinzler.github.io/proj…
Given calibrated multi-view images, MATCH infers static Gaussian splat textures in correspondence across subjects and expressions. This speeds up avatar optimization by a factor of 10x.🧵👇
Guoxing Sun retweeted
🚀 Fast SAM 3D Body — accelerating SAM 3D Body for real-time human mesh recovery!
⚡ 10.25× faster 3D body estimation
⚡ 10,426× faster MHR→SMPL conversion
⏱️ ~65ms end-to-end
🤖 Deployable on humanoid robots
#Robotics
Guoxing Sun retweeted
🔥 #3DV2026 Oral: We present a matrix-free Levenberg-Marquardt optimizer that makes second-order optimization practical for 3DGS! 🚀
Guoxing Sun retweeted
🎾Introducing LATENT: Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data
Dynamic movements, agile whole-body coordination, and rapid reactions. A step toward athletic humanoid sports skills.
Project: zzk273.github.io/LATENT/
Code: github.com/GalaxyGeneralRobo…
Guoxing Sun retweeted
We also release the code and data to run CARI4D on videos from video generation models, checkout at: github.com/NVlabs/CARI4D/blo…
Here are some example results:
Excited to share that CARI4D has been accepted to #CVPR2026 @CVPR
And our code is also released at: github.com/NVlabs/CARI4D
Welcome to play with it and share your feedback🤗!
![GitHub - NVlabs/CARI4D: [CVPR 2026] CARI4D: Category Agnostic 4D Reconstruction of Human-Object...](https://pbs.twimg.com/card_img/2115248817113071617/te8HUmLg?format=jpg&name=800x419)