{
  "id": "1707.06923",
  "version": "v1",
  "published": "2017-07-21T14:45:48.000Z",
  "updated": "2017-07-21T14:45:48.000Z",
  "title": "Pillar Networks for action recognition",
  "authors": [
    "Biswa Sengupta",
    "Yu Qian"
  ],
  "categories": [
    "cs.CV",
    "stat.ML"
  ],
  "abstract": "Image understanding using deep convolutional network has reached human-level performance, yet a closely related problem of video understanding especially, action recognition has not reached the requisite level of maturity. We combine multi-kernels based support-vector-machines (SVM) with a multi-stream deep convolutional neural network to achieve close to state-of-the-art performance on a 51-class activity recognition problem (HMDB-51 dataset); this specific dataset has proved to be particularly challenging for deep neural networks due to the heterogeneity in camera viewpoints, video quality, etc. The resulting architecture is named pillar networks as each (very) deep neural network acts as a pillar for the hierarchical classifiers.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2017-07-21T14:45:48.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "action recognition",
      "pillar networks",
      "multi-stream deep convolutional neural network",
      "deep neural network acts",
      "deep convolutional network"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}