{
  "id": 5218,
  "title": "Question about the format of the code to be submitted",
  "url": "/competitions/multi-modal-gesture-recognition/discussion/5218",
  "author_name": "kongwah",
  "post_date": "2013-07-28T15:41:59.393000",
  "votes": 0,
  "comment_count": 0,
  "views": 525,
  "content": "<p>We want to ask about the code we are supposed to submit, alongside with the prediction on the final evaluation data, by Aug 20.</p>\n<p>Are we supposed to include code for feature extraction, model construction, etc?</p>\n<p>Note that our Linux code is working in the video/image domain. It uses some 3rd party code to extract visual features, e.g. OpenCV, imagemagick, ffmpeg, K-means. The feature extraction is very compute intensive, and uses mpi to expedite the processing. Replicating this on the organizers' platform is of course possible, but this may be more involved.</p>\n<p>&nbsp;</p>\n<p>Telepoints</p>",
  "messages": [
    {
      "id": 27731,
      "postDate": "2013-07-28T15:41:59.393Z",
      "content": "<p>We want to ask about the code we are supposed to submit, alongside with the prediction on the final evaluation data, by Aug 20.</p>\n<p>Are we supposed to include code for feature extraction, model construction, etc?</p>\n<p>Note that our Linux code is working in the video/image domain. It uses some 3rd party code to extract visual features, e.g. OpenCV, imagemagick, ffmpeg, K-means. The feature extraction is very compute intensive, and uses mpi to expedite the processing. Replicating this on the organizers' platform is of course possible, but this may be more involved.</p>\n<p>&nbsp;</p>\n<p>Telepoints</p>"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "27731": ""
  }
}