Chun-Peng Chang

me.jpg

My name is Chun-Peng Chang from Taiwan. I hold a Master’s degree in Media Informatics from RWTH Aachen and completed my Master’s thesis at the Computer Vision Lab of the University of Bonn under the supervision of Prof. Jürgen Gall. I am currently a ELLIS PhD candidate at TU Delft (Cognitive Robotics) with Dr. Holger Caesar, focusing on foundation models in machine perception. Previously, I worked as a researcher at DFKI in Kaiserslautern, Germany, with Dr. Alain Pagani, where I developed algorithms for real-time scene understanding and visual-textual reasoning to improve human-robot collaboration in industrial environments.

In addition to my academic work, I have gained practical experience through internships and research collaborations with companies such as Tomofun, IBM, and Mercedes-Benz.

Timeline of my journey

news

Oct 08, 2025 Listed as an Outstanding Reviewer for ICCV 2025!
Jan 28, 2025 One paper is accepted to ICRA 2025! Link
Oct 28, 2024 One paper is accepted to WACV 2025! Link
Jun 01, 2024 I will participate CVPR 24 in Seattle!
Feb 26, 2024 One paper is accepted to CVPR 2024! Link

selected publications

  1. 3D Spatial Understanding in MLLMs: Disambiguation and Evaluation
    Chun-Peng Chang, Alain Pagani, and Didier Stricker
    In 2025 International Conference on Robotics and Automation (ICRA), 2025
  2. Uni-SLAM: Uncertainty-Aware Neural Implicit SLAM for Real-Time Dense Indoor Scene Reconstruction
    In Proceedings of the Winter Conference on Applications of Computer Vision (WACV), 2025
  3. MiKASA: Multi-Key-Anchor & Scene-Aware Transformer for 3D Visual Grounding
    Chun-Peng Chang, Shaoxiang WangAlain Pagani, and Didier Stricker
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024