QueryProp: Object Query Propagation for High-Performance Video Object Detection

Video object detection has been an important yet challenging topic in computer vision. Traditional methods mainly focus on designing the image-level or box-level feature propagation strategies to exploit temporal information. This paper argues that with a more effective and efficient framework, video detectors can gain improvement terms of both accuracy speed. For this purpose, studies object-level propagation, proposes query (QueryProp) framework for high-performance detection. The proposed QueryProp contains two strategies: 1) is performed from sparse key frames dense non-key reduce redundant computation frames; 2) previous current frame improve representation by context modeling. To further facilitate adaptive gate designed achieve flexible selection. We conduct extensive experiments ImageNet VID dataset. achieves comparable state-of-the-art strikes decent accuracy/speed trade-off. Code available at https://github.com/hf1995/QueryProp.