Flow-Guided Feature Aggregation for Video Object Detection
This work addresses the challenge of accurate object detection in videos for computer vision applications, representing an incremental improvement by enhancing feature aggregation rather than introducing a new paradigm.
The paper tackles the problem of video object detection by addressing degenerated object appearances like motion blur, proposing an end-to-end flow-guided feature aggregation framework that leverages temporal coherence on the feature level. It significantly improves accuracy over single-frame baselines in ImageNet VID, especially for fast-moving objects, and powered the winning entry in the 2017 challenge.
Extending state-of-the-art object detectors from image to video is challenging. The accuracy of detection suffers from degenerated object appearances in videos, e.g., motion blur, video defocus, rare poses, etc. Existing work attempts to exploit temporal information on box level, but such methods are not trained end-to-end. We present flow-guided feature aggregation, an accurate and end-to-end learning framework for video object detection. It leverages temporal coherence on feature level instead. It improves the per-frame features by aggregation of nearby features along the motion paths, and thus improves the video recognition accuracy. Our method significantly improves upon strong single-frame baselines in ImageNet VID, especially for more challenging fast moving objects. Our framework is principled, and on par with the best engineered systems winning the ImageNet VID challenges 2016, without additional bells-and-whistles. The proposed method, together with Deep Feature Flow, powered the winning entry of ImageNet VID challenges 2017. The code is available at https://github.com/msracver/Flow-Guided-Feature-Aggregation.