.png){kind=icon}
Return
Angle-aware multi-object fish tracking with unique forward orientation estimation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.aquaeng.2025.102658.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes AOMO-FishTracker: an angle-oriented two-stage fish tracking framework. • Introduces oriented bounding boxes with forward orientation representation. • Designs RIoU-based matching and angle-difference filtering for robust tracking. • Improves IDF1 by 2.72% and MOTA by 1.53% on challenging dense fish datasets.
Journal
IF:
4.3
Papers:
1.6K
Citations:
4.2K
Organization
Cited Papers
No cited papers available