{
  "id": 165393,
  "title": "can anyone kindly describe about the bbox column?",
  "url": "/competitions/global-wheat-detection/discussion/165393",
  "author_name": "Pinaki Sen",
  "post_date": "2020-07-09T14:36:43.178000",
  "votes": 2,
  "comment_count": 6,
  "views": 0,
  "content": "<p>It will be really helpful if someone kindly describe all four numeric values of the bbox column.</p>",
  "messages": [
    {
      "id": 921902,
      "postDate": "2020-07-09T16:33:32.713Z",
      "content": "<p>first two numbers are (column, row) coordinates of the bbox's upper left corner\n3rd number: width of the bbox\n4th number: height of the bbox\nall numbers are in pixels</p>",
      "rawMarkdown": "first two numbers are (column, row) coordinates of the bbox's upper left corner\n3rd number: width of the bbox\n4th number: height of the bbox\nall numbers are in pixels\n\n",
      "votes": 2,
      "replies": [
        {
          "id": 922161,
          "postDate": "2020-07-09T20:42:45.660Z",
          "content": "<p>Thank you.</p>",
          "rawMarkdown": "Thank you.",
          "votes": 1
        }
      ]
    },
    {
      "id": 921770,
      "postDate": "2020-07-09T14:36:43.177Z",
      "content": "<p>It will be really helpful if someone kindly describe all four numeric values of the bbox column.</p>",
      "rawMarkdown": "It will be really helpful if someone kindly describe all four numeric values of the bbox column.",
      "votes": 2
    },
    {
      "id": 922100,
      "postDate": "2020-07-09T20:01:20.653Z",
      "content": "<p>Hi <a href=\"/senpinaki222\">@senpinaki222</a>, you can find your answer <a href=\"https://www.kaggle.com/c/global-wheat-detection/data\">here</a>:</p>\n\n<p>We are using Pascal VOC format, so  [xmin, ymin, width, height], using pixel values.\nxmin,ymin is the top left corner, while adding width and heigth you will get the bottom right corner of the box.</p>\n\n<p>In general about the possbile, most common formats:\nCOCO Bounding box: (x-top left, y-top left, width, height)\nPascal VOC Bounding box :(x-top left, y-top left,x-bottom right, y-bottom right) </p>\n\n<p>Some more info about how to use and especially convert the numbers, you can find <a href=\"https://www.kaggle.com/chrisden/end2end-object-detection-with-fasterrcnn\">here</a>. Let me know if you have questions and welcome to this competition!</p>",
      "rawMarkdown": "Hi @senpinaki222, you can find your answer [here](https://www.kaggle.com/c/global-wheat-detection/data):\n\nWe are using Pascal VOC format, so  [xmin, ymin, width, height], using pixel values.\nxmin,ymin is the top left corner, while adding width and heigth you will get the bottom right corner of the box.\n\nIn general about the possbile, most common formats:\nCOCO Bounding box: (x-top left, y-top left, width, height)\nPascal VOC Bounding box :(x-top left, y-top left,x-bottom right, y-bottom right) \n\nSome more info about how to use and especially convert the numbers, you can find [here](https://www.kaggle.com/chrisden/end2end-object-detection-with-fasterrcnn). Let me know if you have questions and welcome to this competition!\n",
      "votes": 2,
      "replies": [
        {
          "id": 922160,
          "postDate": "2020-07-09T20:42:21.353Z",
          "content": "<p>Thank you. The description is really helpful.</p>",
          "rawMarkdown": "Thank you. The description is really helpful.",
          "votes": 1
        },
        {
          "id": 939877,
          "postDate": "2020-07-22T14:48:21.020Z",
          "rawMarkdown": "",
          "isDeleted": true
        },
        {
          "id": 942881,
          "postDate": "2020-07-24T03:43:13.957Z",
          "content": "<p><a href=\"/x2t2cx\">@x2t2cx</a> \nIt's simple:\n[x0, y0, width, height] --&gt; [x0, y0, x1, y1]\nx1 = x0 + width\ny1 = y0 + height</p>",
          "rawMarkdown": "@x2t2cx \nIt's simple:\n[x0, y0, width, height] --&gt; [x0, y0, x1, y1]\nx1 = x0 + width\ny1 = y0 + height"
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 921902,
      "author_name": "zastaph",
      "author_url": "",
      "post_date": "2020-07-09T16:33:32.713000",
      "content": "<p>first two numbers are (column, row) coordinates of the bbox's upper left corner\n3rd number: width of the bbox\n4th number: height of the bbox\nall numbers are in pixels</p>",
      "votes": 2,
      "replies": [
        {
          "id": 922161,
          "author_name": "Pinaki Sen",
          "author_url": "",
          "post_date": "2020-07-09T20:42:45.660000",
          "content": "<p>Thank you.</p>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 922100,
      "author_name": "from coffee import *",
      "author_url": "",
      "post_date": "2020-07-09T20:01:20.653000",
      "content": "<p>Hi <a href=\"/senpinaki222\">@senpinaki222</a>, you can find your answer <a href=\"https://www.kaggle.com/c/global-wheat-detection/data\">here</a>:</p>\n\n<p>We are using Pascal VOC format, so  [xmin, ymin, width, height], using pixel values.\nxmin,ymin is the top left corner, while adding width and heigth you will get the bottom right corner of the box.</p>\n\n<p>In general about the possbile, most common formats:\nCOCO Bounding box: (x-top left, y-top left, width, height)\nPascal VOC Bounding box :(x-top left, y-top left,x-bottom right, y-bottom right) </p>\n\n<p>Some more info about how to use and especially convert the numbers, you can find <a href=\"https://www.kaggle.com/chrisden/end2end-object-detection-with-fasterrcnn\">here</a>. Let me know if you have questions and welcome to this competition!</p>",
      "votes": 2,
      "replies": [
        {
          "id": 922160,
          "author_name": "Pinaki Sen",
          "author_url": "",
          "post_date": "2020-07-09T20:42:21.353000",
          "content": "<p>Thank you. The description is really helpful.</p>",
          "votes": 1,
          "replies": []
        },
        {
          "id": 939877,
          "author_name": "",
          "author_url": "",
          "post_date": "2020-07-22T14:48:21.020000",
          "content": "",
          "votes": 0,
          "replies": []
        },
        {
          "id": 942881,
          "author_name": "Matt",
          "author_url": "",
          "post_date": "2020-07-24T03:43:13.957000",
          "content": "<p><a href=\"/x2t2cx\">@x2t2cx</a> \nIt's simple:\n[x0, y0, width, height] --&gt; [x0, y0, x1, y1]\nx1 = x0 + width\ny1 = y0 + height</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "921902": "first two numbers are (column, row) coordinates of the bbox's upper left corner\n3rd number: width of the bbox\n4th number: height of the bbox\nall numbers are in pixels\n\n",
    "921770": "It will be really helpful if someone kindly describe all four numeric values of the bbox column.",
    "922100": "Hi @senpinaki222, you can find your answer [here](https://www.kaggle.com/c/global-wheat-detection/data):\n\nWe are using Pascal VOC format, so  [xmin, ymin, width, height], using pixel values.\nxmin,ymin is the top left corner, while adding width and heigth you will get the bottom right corner of the box.\n\nIn general about the possbile, most common formats:\nCOCO Bounding box: (x-top left, y-top left, width, height)\nPascal VOC Bounding box :(x-top left, y-top left,x-bottom right, y-bottom right) \n\nSome more info about how to use and especially convert the numbers, you can find [here](https://www.kaggle.com/chrisden/end2end-object-detection-with-fasterrcnn). Let me know if you have questions and welcome to this competition!\n"
  }
}