- Calibration-Free Multi-Object Tracking via Scale-Invariant Topological Depth and Adaptive Uncertainty Gating

Jin Hong, Yoojin Han, Junseok Kwon, IEEE International Conference on Advanced Visual and Signal-Based Systems (AVSS), Lecce, Italy, 2026

ACK: RS-2021II211341

- Text-Centric Multimodal Alignment via Dual-Level Optimization

Jin Hong, Juhyeon Park, and Junseok Kwon, The 40th International Conference on Information Networking (ICOIN), Hanoi, Vietnam, 2026 [pdf] 

ACK: RS-2025-02217071

- Occlusion-Robust Multi-Object Tracking with Adaptive Feature Management and Motion Compensation

Jin Hong, Yoojin Han, Junseok Kwon, IEEE International Conference on Advanced Visual and Signal-Based Systems (AVSS), Tainan, Taiwan, 2025  [pdf] [code]

[Excellent Paper Award]

ACK: 2021-0-01341, Artificial Intelligence Graduate School Program

- diffMusic: A Multi-Stage Model for Contextually Aligned Music Generation from Visual Inputs 

Jin Hong, Taeun Lee, and Junseok Kwon , 14th Asian Conference on Computer Vision (ACCV) Workshop, Oral presentation, Hanoi, Vietnam, 2024 

- DiffMusic: Efficient Music Generation From a Single Image Using Diffusion-Based Representations

Jin Hong, Juhyeon Park, and Junseok Kwon, IEEE Transactions on Audio, Speech and Language Processing (T-ASLP), vol. 34, pp. 1126-1136, 2026  [pdf] [code]

ACK: 2021-0-01341, Artificial Intelligence Graduate School Program, Chung-Ang University Graduate Research Scholarship

- End-to-End Metric Learning From Corrupted Images Using Triplet Dimensionality Reduction Loss

Juhyeon Park, Jin Hong,  and Junseok Kwon, Expert Systems with Applications (ESWA),  vol 238, Part C, 2024 [pdf] [code]

ACK: 2021-0-01341, Artificial Intelligence Graduate School Program, NRF-2020R1C1C1004907