“MaskControl: Spatio-Temporal Control for Masked Motion Synthesis” selected as 🏆 Award Candidate at ICCV 2025! 🌺

Our paper “MaskControl: Spatio-Temporal Control for Masked Motion Synthesis” has been selected as an 🏆 Award Candidate at ICCV 2025! 🌺✨
It’s a huge honor to see our work recognized among the top papers this year.

ICCV is one of the world’s premier conferences in AI and computer vision!
⭐ 11,239 paper submissions
⭐ 2,701 accepted papers (24% acceptance rate)
⭐ 263 highlights
⭐ 64 oral presentations
⭐ 26 award candidates

______________________________________________________________
MaskControl is the first to bring controllability to Masked Motion Models via two novel components:

🧠 Logits Regularizer – ControlNet-like for Masked Models
🎯 Logits Optimization – Inference-time guidance tailored for Masked Models

To address the non-differentiable nature of quantized models, we propose Differentiable Expectation Sampling, which relaxes quantization and enables conversion across multiple representation spaces.

__________________________________________________________________
🔗 Project Page: https://lnkd.in/erRXcgrG
💻 Code: https://lnkd.in/eET4XnTd
📄 Paper: https://lnkd.in/e_Vpp4PB
🎥 YouTube: https://lnkd.in/e3S68tsz

Wednesday, Oct 22, 2025 Oral Session
8:00-9:15 Oral Session 3B: Human Modeling (Kalakaua Ballroom)
10:45 – 12:45 Poster Session 3 & Exhibit Hall (Exhibit Hall I)
Paper ID: 2360

__________________________________________________________________
Big thanks to all my amazing collaborators Muhammad Usama Saleem, Korrawe Karunratanakul, Pu Wang, Chen Chen, Hongfei Xue, Chuan Guo, Junli Cao, Jian Ren, Sergey Tulyakov