{
  "id": 130762,
  "title": "Can anyone share face detected bbox's?",
  "url": "/competitions/deepfake-detection-challenge/discussion/130762",
  "author_name": "",
  "post_date": "2020-02-16T07:13:30.604092500Z",
  "votes": null,
  "comment_count": 6,
  "views": 0,
  "content": "<p>I would appreciate you to share face detection coordinates for whole dataset. Any format you like.</p>",
  "messages": [
    {
      "id": "747258",
      "postDate": "02/16/2020 07:13:30",
      "content": "<p>I would appreciate you to share face detection coordinates for whole dataset. Any format you like.</p>",
      "rawMarkdown": "I would appreciate you to share face detection coordinates for whole dataset. Any format you like.",
      "votes": null
    },
    {
      "id": "747381",
      "postDate": "02/16/2020 11:09:04",
      "content": "<p>Why don't you do it by yourself? There are plenty of public code doing this easily.</p>",
      "rawMarkdown": "Why don't you do it by yourself? There are plenty of public code doing this easily.",
      "votes": null
    },
    {
      "id": "747477",
      "postDate": "02/16/2020 13:41:29",
      "content": "<p>Yes, I know. I just thought to make my start faster.</p>",
      "rawMarkdown": "Yes, I know. I just thought to make my start faster.",
      "votes": null
    },
    {
      "id": "747715",
      "postDate": "02/16/2020 18:58:01",
      "content": "<p>I'd love to share. But I saved my results as cropped face images. I don't want to waste $100 aws credits and ~2weeks time to do it again.</p>",
      "rawMarkdown": "I'd love to share. But I saved my results as cropped face images. I don't want to waste $100 aws credits and ~2weeks time to do it again.",
      "votes": null
    },
    {
      "id": "747743",
      "postDate": "02/16/2020 19:37:13",
      "content": "<p>Even if you had face bboxes for the training set, you'd still need to come up with a way to detect faces in the final test set yourself.</p>",
      "rawMarkdown": "Even if you had face bboxes for the training set, you'd still need to come up with a way to detect faces in the final test set yourself.",
      "votes": null
    },
    {
      "id": "754724",
      "postDate": "02/24/2020 01:17:53",
      "content": "<p><a href=\"/unkownhihi\">@unkownhihi</a> That is monstrous! 2 weeks time to make a dataset? What if you decided that there is a better way to make the dataset???</p>",
      "rawMarkdown": "unkownhihi That is monstrous! 2 weeks time to make a dataset? What if you decided that there is a better way to make the dataset???",
      "votes": null
    },
    {
      "id": "767616",
      "postDate": "03/09/2020 22:24:27",
      "content": "<p><a href=\"/roguekk007\">@roguekk007</a> I didn't run like 24/7 2 weeks to make this dataset. I run it whenever I have time(around 3-4 hours per day and sometimes can't run at all).</p>",
      "rawMarkdown": "roguekk007 I didn't run like 24/7 2 weeks to make this dataset. I run it whenever I have time(around 3-4 hours per day and sometimes can't run at all).",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 747381,
      "author_name": "khahuras",
      "author_url": "",
      "post_date": "02/16/2020 11:09:04",
      "content": "<p>Why don't you do it by yourself? There are plenty of public code doing this easily.</p>",
      "votes": null,
      "replies": [
        {
          "id": 747477,
          "author_name": "lavrikovav",
          "author_url": "",
          "post_date": "02/16/2020 13:41:29",
          "content": "<p>Yes, I know. I just thought to make my start faster.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 747715,
      "author_name": "unkownhihi",
      "author_url": "",
      "post_date": "02/16/2020 18:58:01",
      "content": "<p>I'd love to share. But I saved my results as cropped face images. I don't want to waste $100 aws credits and ~2weeks time to do it again.</p>",
      "votes": null,
      "replies": [
        {
          "id": 754724,
          "author_name": "roguekk007",
          "author_url": "",
          "post_date": "02/24/2020 01:17:53",
          "content": "<p><a href=\"/unkownhihi\">@unkownhihi</a> That is monstrous! 2 weeks time to make a dataset? What if you decided that there is a better way to make the dataset???</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 767616,
          "author_name": "unkownhihi",
          "author_url": "",
          "post_date": "03/09/2020 22:24:27",
          "content": "<p><a href=\"/roguekk007\">@roguekk007</a> I didn't run like 24/7 2 weeks to make this dataset. I run it whenever I have time(around 3-4 hours per day and sometimes can't run at all).</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 747743,
      "author_name": "humananalog",
      "author_url": "",
      "post_date": "02/16/2020 19:37:13",
      "content": "<p>Even if you had face bboxes for the training set, you'd still need to come up with a way to detect faces in the final test set yourself.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "747258": "I would appreciate you to share face detection coordinates for whole dataset. Any format you like.",
    "747381": "Why don't you do it by yourself? There are plenty of public code doing this easily.",
    "747477": "Yes, I know. I just thought to make my start faster.",
    "747715": "I'd love to share. But I saved my results as cropped face images. I don't want to waste $100 aws credits and ~2weeks time to do it again.",
    "747743": "Even if you had face bboxes for the training set, you'd still need to come up with a way to detect faces in the final test set yourself.",
    "754724": "unkownhihi That is monstrous! 2 weeks time to make a dataset? What if you decided that there is a better way to make the dataset???",
    "767616": "roguekk007 I didn't run like 24/7 2 weeks to make this dataset. I run it whenever I have time(around 3-4 hours per day and sometimes can't run at all)."
  },
  "source": "meta"
}