Mingrun Lin is a master's researcher focusing on multi-object tracking and underwater computer vision. His research interests include underwater multi-fish tracking, scene-aware representation learning, and spatio-temporal modeling for complex aquatic environments. His recent work investigates robust tracking methods under challenging conditions such as low illumination, water surface reflections, and dense fish interactions. He has developed adaptive feature calibration modules and trajectory-level temporal modeling frameworks to improve tracking stability and identity preservation across diverse underwater and above-water scenarios. He has built real-world multi-fish tracking datasets collected from aquaculture environments and is interested in applying intelligent perception technologies to precision aquaculture and aquatic behavior analysis.