Nearshore optical video object detector based on temporal branch and spatial feature enhancement

Yuanlin Zhao, Wei Li, Jiangang Ding, Yansong Wang, Lili Pei, Aojia Tian

Abstract: Highlights•We present a novel pipeline for detecting nearshore objects, called NVID.•YCLM precisely extracts correlated frame temporal information.•PDA acquires object spatial characteristics with minimal computational cost.•FRENet achieves efficient detection through fast re-parameterization.•Post-TensorRT quantization, NVIDIA is deployable on Jetson Nano in real-time.