Ling-Hao (Evan) CHEN
@Evan_THU
Followers
350
Following
736
Media
56
Statuses
556
Ph.D. student @Tsinghua_Uni, Intern @IDEACVR, Vision, Graphics, and Motion Synthesis! ex: @XDUofChina
HongKong, P.R. China
Joined February 2022
@SIGGRAPHAsia 2025, we introduce a lightweight motion retargeting (transfer) system, Motion2Motion. 🔥Beyond deep learning, Motion2Motion is a motion-matching-based system, which runs on CPU-only devices in REAL time. See more at https://t.co/8ONOhywta5
#SIGGRAPH
1
3
11
🌟Motion2Motion supports diverse applications: SMPL-to-any character retargeting, cross-species motion transfer, key-frame transfer... Motion2Motion Blender add-on is also developed for industrial usage.
0
0
0
🚨SIGGRAPH Asia 2025 Paper Alert 🚨 ➡️Paper Title: Motion2Motion: Cross-topology Motion Transfer with Sparse Correspondence 🌟Few pointers from the paper 🎯This work studies the challenge of transfer animations between characters whose skeletal topologies differ substantially.
0
4
18
Happy birthday to the finest father — whose grace in triumph, indefatigable strength in adversity, and quiet courage to rise above the fray continue to guide me everyday!
732
100
2K
@docmilanfar Computer vision was driven by machine learning long before 2012. What happened with AlexNet was a shift to deep neural networks as the dominant form of ML. And, with this, came the focus on training data at scale. But this didn't just transform CV. It transformed everything.
4
3
71
#ICML2025
https://t.co/aM98fshdBh We surveyed 100+ courses across 19 departments at Tsinghua University. With expert and model filtering, we curated a graduate-level, Olympiad-difficulty, multi-disciplinary benchmark R-Bench. Even GPT-4o struggles (33.4% on multimodal)!
1
5
19
🥳Excited to share our work #FullDiT: Multi-Task Video Generative Foundation Model with Full Attention https://t.co/aExynj80Hn, a unified foundation model for video generation that seamlessly integrates multiple conditions via unified full-attention mechanisms.
huggingface.co
0
8
60
amazing
This year, an important mission is to let simulated humanoids solve long-term manipulation tasks in challenging dynamic, cluttered environments. This is a simple demonstration achieved by TokenHSI (CVPR 2025). The humanoid here is controlled by a single, unified model.
0
0
2
A concise and comprehensive survey of 2D, 3D, and 4D generation.
0
16
89
Lost in various AI generations? Check out this survey. The first attempt to systematically unify the study of 2D, 2D+Time(video), 3D, and (3D+Time)4D generation within a single framework. Arxiv: https://t.co/MEkMO4bUbm All relevant links: https://t.co/vqj104c5Ax Simulating the
0
2
8
thanks for sharing our work❤️
🚨CVPR 2025 Paper Alert 🚨 ➡️Paper Title: HumanMM: Global Human Motion Recovery from Multi-shot Videos 🌟Few pointers from the paper 🎯In this paper, authors have presented a novel framework designed to reconstruct long-sequence 3D human motion in the world coordinates from
1
0
5
Project: https://t.co/hV20cNQufz Paper: https://t.co/qN0efxaIOB Demo: https://t.co/IkPGUMY11x ❤️Work mainly done by @YuhongZhan31576 @GuanlinWu0930 when their internship at @IDEACVR . Happy to contribute to this project with all coauthors jointly. #pingpong #Olympics #cvpr
0
0
3
🚩 Lifting from 2D to video/to 3D, to 4D. This is the latest survey about the world simulator.
🚀 How can we bridge the gap between 2D, video, 3D, and 4D generation for real-world simulation? Our new survey presents the first unified framework for multimodal generative models! ✨ GitHub: https://t.co/KT0DZkmsrg arXiv: https://t.co/Zvx3V5Op4S
#GenerativeAI #survey
0
0
3
incredible
0
0
1
Interested in: ✅Humanoids mastering scalable motor skills for everyday interactions ✅Whole-body loco-manipulation w/ diverse tasks and objects ✅Physically plausible HOI animation Meet InterMimic #CVPR2025 ArXiv: https://t.co/GvsfJ0y8b2 Project: https://t.co/6onZLPp7bx 🧵[1/9]
13
46
266
早上起床看到这个 真的逆天
人类外交历史上的惊人一幕，美国白宫上演现实版纸牌屋： 美国政府总统和来访的乌克兰总统泽伦斯基发生争吵，最后，川普和万斯以绝对优势获得2:0的胜利，将泽伦斯基赶出了白宫。
0
0
0
Input a video to generate 3D dance animations in real time and connect to UE5 for anime-style animation output. The free version 1.6.2 is now available!
5
32
207