.png){kind=icon}
Return
Object-state-based frame sampling for efficient video object detection
.png){kind=icon}
OA
AI
DOI：10.1016/j.heliyon.2026.e45426.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• An object-state-based frame sampling is proposed for video object detection.
• IoU-based anchors preserve key object-state transitions during sampling.
• KITTI and Kgalagadi confirm detector-signal alignment under controlled budgets.
• Compression behavior differs by video regimes, and cross-domain fusion separately expands appearance diversity.
Keywords:
Frame sampling
Data fusion
Object detection
Data-centric
Object-state-based curation
Journal
IF:
3.6
Papers:
3.8W
Citations:
10.5W
Organization
No organization information available
Cited Papers
No cited papers available