RN-VID: A Feature Fusion Architecture for Video Object Detection
This work addresses efficiency and accuracy in video object detection, which is important for applications like surveillance and autonomous driving, but it appears incremental as it builds on existing single-frame detectors.
The paper tackles video object detection by proposing RN-VID, a feature fusion architecture that reuses information across frames to enhance feature maps, achieving better mean average precision (mAP) than single-frame detectors with minimal inference cost.
Consecutive frames in a video are highly redundant. Therefore, to perform the task of video object detection, executing single frame detectors on every frame without reusing any information is quite wasteful. It is with this idea in mind that we propose RN-VID (standing for RetinaNet-VIDeo), a novel approach to video object detection. Our contributions are twofold. First, we propose a new architecture that allows the usage of information from nearby frames to enhance feature maps. Second, we propose a novel module to merge feature maps of same dimensions using re-ordering of channels and 1 x 1 convolutions. We then demonstrate that RN-VID achieves better mean average precision (mAP) than corresponding single frame detectors with little additional cost during inference.