{
  "id": 121434,
  "title": "3d convolutions vs 2d convolutions",
  "url": "/competitions/deepfake-detection-challenge/discussion/121434",
  "author_name": "",
  "post_date": "2019-12-13T09:37:25.894921500Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Does anyone have any idea of resources/information regarding the effectiveness of 3d convolutions of successive frames vs 2d conv for single images? I was thinking that giving the model multiple frames would allow for comparisons and the ability to detect unnatural distortions over time. </p>",
  "messages": [
    {
      "id": "694196",
      "postDate": "12/13/2019 09:37:25",
      "content": "<p>Does anyone have any idea of resources/information regarding the effectiveness of 3d convolutions of successive frames vs 2d conv for single images? I was thinking that giving the model multiple frames would allow for comparisons and the ability to detect unnatural distortions over time. </p>",
      "rawMarkdown": "Does anyone have any idea of resources/information regarding the effectiveness of 3d convolutions of successive frames vs 2d conv for single images? I was thinking that giving the model multiple frames would allow for comparisons and the ability to detect unnatural distortions over time.",
      "votes": null
    },
    {
      "id": "694243",
      "postDate": "12/13/2019 10:34:41",
      "content": "<p><a href=\"http://blog.qure.ai/notes/deep-learning-for-videos-action-recognition-review\">http://blog.qure.ai/notes/deep-learning-for-videos-action-recognition-review</a></p>\n\n<p>You will probably find this write-up useful. </p>",
      "rawMarkdown": "http://blog.qure.ai/notes/deep-learning-for-videos-action-recognition-review\n\nYou will probably find this write-up useful.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 694243,
      "author_name": "ryches",
      "author_url": "",
      "post_date": "12/13/2019 10:34:41",
      "content": "<p><a href=\"http://blog.qure.ai/notes/deep-learning-for-videos-action-recognition-review\">http://blog.qure.ai/notes/deep-learning-for-videos-action-recognition-review</a></p>\n\n<p>You will probably find this write-up useful. </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "694196": "Does anyone have any idea of resources/information regarding the effectiveness of 3d convolutions of successive frames vs 2d conv for single images? I was thinking that giving the model multiple frames would allow for comparisons and the ability to detect unnatural distortions over time.",
    "694243": "http://blog.qure.ai/notes/deep-learning-for-videos-action-recognition-review\n\nYou will probably find this write-up useful."
  },
  "source": "meta"
}