VISION & LEARNING LABORATORY @ CAU
The laboratory was formed in Mar. 2020 and is led by Prof. Eunwoo Kim.
The aim of VISION & LEARNING LABORATORY at Chung-Ang University is to push the boundary of machine learning research by developing efficient, versatile, and optimal machine learning models, towards general-purpose artificial intelligence. We engage in research to explore methods that understand and learn any intellectual task that human beings can do.
Our research interests include efficient machine learning, automated machine learning, multi-modal learning, multi-task learning, continual learning, machine unlearning, representation learning, and their applications to computer vision and robotics, but not limited to.
NEWS
- [09/2025] A paper on in-context learning agents is accepted to CoRL 2025 Workshop.
- [09/2025] A paper on cross-modal learning is accepted to Neural Networks.
- [06/2025] Two papers on continual learning are accepted to ICCV 2025.
- [06/2025] A paper on text-video retrieval is accepted to Pattern Recognition Letters.
- [05/2025] A paper on 3D human pose estimation is accepted to CVIU.
- [03/2025] A paper on efficient continual learning is accepted to Neural Networks.
- [02/2025] A paper on multi-modal learning is accepted to CVPR 2025 (Highlight).
- [01/2025] A paper on self-corrective robot task planning is accepted to ICRA 2025.
- [12/2024] A paper on dataset condensation is accepted to Pattern Recognition Letters.
- [11/2024] A paper on neural radiance fields is accepted to IEEE Signal Processing Letters.
- [10/2024] A paper on continual learning is accepted to NeurIPS 2024 Workshop.
- [07/2024] A paper on active learning is accepted to IEEE Signal Processing Letters.
- [06/2024] A paper on task planning based on LLMs is accepted to IROS 2024.
- [06/2024] A paper on self-supervised learning is accepted to Pattern Recognition Letters.
- [04/2024] A paper on neural architecture search is accepted to IEEE Trans. on Image Processing.
- [12/2023] A paper on high-dynamic range imaging is accepted to ICASSP 2024.
- [07/2023] A paper on generation-based continual learning is accepted to ICCV 2023.