{
  "id": "2009.09660",
  "version": "v1",
  "published": "2020-09-21T07:55:50.000Z",
  "updated": "2020-09-21T07:55:50.000Z",
  "title": "Feature Flow: In-network Feature Flow Estimation for Video Object Detection",
  "authors": [
    "Ruibing Jin",
    "Guosheng Lin",
    "Changyun Wen",
    "Jianliang Wang",
    "Fayao Liu"
  ],
  "categories": [
    "cs.CV"
  ],
  "abstract": "Optical flow, which expresses pixel displacement, is widely used in many computer vision tasks to provide pixel-level motion information. However, with the remarkable progress of the convolutional neural network, recent state-of-the-art approaches are proposed to solve problems directly on feature-level. Since the displacement of feature vector is not consistent to the pixel displacement, a common approach is to:forward optical flow to a neural network and fine-tune this network on the task dataset. With this method,they expect the fine-tuned network to produce tensors encoding feature-level motion information. In this paper, we rethink this de facto paradigm and analyze its drawbacks in the video object detection task. To mitigate these issues, we propose a novel network (IFF-Net) with an \\textbf{I}n-network \\textbf{F}eature \\textbf{F}low estimation module (IFF module) for video object detection. Without resorting pre-training on any additional dataset, our IFF module is able to directly produce \\textbf{feature flow} which indicates the feature displacement. Our IFF module consists of a shallow module, which shares the features with the detection branches. This compact design enables our IFF-Net to accurately detect objects, while maintaining a fast inference speed. Furthermore, we propose a transformation residual loss (TRL) based on \\textit{self-supervision}, which further improves the performance of our IFF-Net. Our IFF-Net outperforms existing methods and sets a state-of-the-art performance on ImageNet VID.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2020-09-21T07:55:50.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "video object detection",
      "in-network feature flow estimation",
      "tensors encoding feature-level motion",
      "encoding feature-level motion information"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}