{
  "id": 304749,
  "title": "Missing Annotations in train images csv",
  "url": "/competitions/tensorflow-great-barrier-reef/discussion/304749",
  "author_name": "",
  "post_date": "2022-02-02T10:58:56.300533100Z",
  "votes": 3,
  "comment_count": 1,
  "views": 0,
  "content": "<p>We have noticed that in some images of the train set some detection boxes are missing, while they should have been.</p>\n<p>This is shown by the fact that the images are frames from videos and are sequential.</p>\n<p>As shown in the images on the Links</p>\n<p><a href=\"https://ibb.co/R2Wvt7D\" target=\"_blank\">https://ibb.co/R2Wvt7D</a><br>\n<a href=\"https://ibb.co/1fKhrW5\" target=\"_blank\">https://ibb.co/1fKhrW5</a><br>\n<a href=\"https://ibb.co/k6k2wx7\" target=\"_blank\">https://ibb.co/k6k2wx7</a></p>",
  "messages": [
    {
      "id": "1672867",
      "postDate": "02/02/2022 10:58:56",
      "content": "<p>We have noticed that in some images of the train set some detection boxes are missing, while they should have been.</p>\n<p>This is shown by the fact that the images are frames from videos and are sequential.</p>\n<p>As shown in the images on the Links</p>\n<p><a href=\"https://ibb.co/R2Wvt7D\" target=\"_blank\">https://ibb.co/R2Wvt7D</a><br>\n<a href=\"https://ibb.co/1fKhrW5\" target=\"_blank\">https://ibb.co/1fKhrW5</a><br>\n<a href=\"https://ibb.co/k6k2wx7\" target=\"_blank\">https://ibb.co/k6k2wx7</a></p>",
      "rawMarkdown": "We have noticed that in some images of the train set some detection boxes are missing, while they should have been.\n\nThis is shown by the fact that the images are frames from videos and are sequential.\n\nAs shown in the images on the Links\n\nhttps://ibb.co/R2Wvt7D\nhttps://ibb.co/1fKhrW5\nhttps://ibb.co/k6k2wx7",
      "votes": null
    },
    {
      "id": "1676548",
      "postDate": "02/05/2022 04:47:40",
      "content": "<p>Hi there I noticed something similar where some COTS are visible in earlier frames before the first bounding box.  I created a <a href=\"https://www.kaggle.com/bartmaciszewski/more-annotations\" target=\"_blank\">notebook to try to annotate</a> those and build a bigger training set.</p>",
      "rawMarkdown": "Hi there I noticed something similar where some COTS are visible in earlier frames before the first bounding box.  I created a [notebook to try to annotate](https://www.kaggle.com/bartmaciszewski/more-annotations) those and build a bigger training set.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1676548,
      "author_name": "bartmaciszewski",
      "author_url": "",
      "post_date": "02/05/2022 04:47:40",
      "content": "<p>Hi there I noticed something similar where some COTS are visible in earlier frames before the first bounding box.  I created a <a href=\"https://www.kaggle.com/bartmaciszewski/more-annotations\" target=\"_blank\">notebook to try to annotate</a> those and build a bigger training set.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1672867": "We have noticed that in some images of the train set some detection boxes are missing, while they should have been.\n\nThis is shown by the fact that the images are frames from videos and are sequential.\n\nAs shown in the images on the Links\n\nhttps://ibb.co/R2Wvt7D\nhttps://ibb.co/1fKhrW5\nhttps://ibb.co/k6k2wx7",
    "1676548": "Hi there I noticed something similar where some COTS are visible in earlier frames before the first bounding box.  I created a [notebook to try to annotate](https://www.kaggle.com/bartmaciszewski/more-annotations) those and build a bigger training set."
  },
  "source": "meta"
}