{
  "id": 304165,
  "title": "Masks dataset for unique COTS",
  "url": "/competitions/tensorflow-great-barrier-reef/discussion/304165",
  "author_name": "",
  "post_date": "2022-01-31T04:14:32.841869100Z",
  "votes": 10,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Using the tracking of unique COTS from <a href=\"https://www.kaggle.com/hengck23\" target=\"_blank\">@hengck23</a> in <a href=\"https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/299356\" target=\"_blank\">https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/299356</a>, I drawed the masks with label-studio of the middle appearance frame for each of the 228 unique COTS. Some of them have a tiny hole inside the mask because of incomplete inpainting. </p>\n<p>In case you want to experiment some adaptation of copy-paste augmentation or any other technique, here is the dataset with the crops of the unique COTS and their associated mask:<br>\n<a href=\"https://www.kaggle.com/alexandrecc/cots-masks\" target=\"_blank\">https://www.kaggle.com/alexandrecc/cots-masks</a></p>",
  "messages": [
    {
      "id": "1669912",
      "postDate": "01/31/2022 04:14:32",
      "content": "<p>Using the tracking of unique COTS from <a href=\"https://www.kaggle.com/hengck23\" target=\"_blank\">@hengck23</a> in <a href=\"https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/299356\" target=\"_blank\">https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/299356</a>, I drawed the masks with label-studio of the middle appearance frame for each of the 228 unique COTS. Some of them have a tiny hole inside the mask because of incomplete inpainting. </p>\n<p>In case you want to experiment some adaptation of copy-paste augmentation or any other technique, here is the dataset with the crops of the unique COTS and their associated mask:<br>\n<a href=\"https://www.kaggle.com/alexandrecc/cots-masks\" target=\"_blank\">https://www.kaggle.com/alexandrecc/cots-masks</a></p>",
      "rawMarkdown": "Using the tracking of unique COTS from @hengck23 in https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/299356, I drawed the masks with label-studio of the middle appearance frame for each of the 228 unique COTS. Some of them have a tiny hole inside the mask because of incomplete inpainting. \n\nIn case you want to experiment some adaptation of copy-paste augmentation or any other technique, here is the dataset with the crops of the unique COTS and their associated mask:\nhttps://www.kaggle.com/alexandrecc/cots-masks",
      "votes": null
    },
    {
      "id": "1669925",
      "postDate": "01/31/2022 04:32:09",
      "content": "<p>Thank you for sharing.</p>",
      "rawMarkdown": "Thank you for sharing.",
      "votes": null
    },
    {
      "id": "1670042",
      "postDate": "01/31/2022 07:04:00",
      "content": "<p>Great! Thanks for sharing👍👍👍</p>",
      "rawMarkdown": "Great! Thanks for sharing👍👍👍",
      "votes": null
    },
    {
      "id": "1670150",
      "postDate": "01/31/2022 09:00:49",
      "content": "<p>That's great. But are people going to use segmentation models? I tried to use them but since it's segmentation model, I couldn't get the bbox coordinates for COTS.</p>",
      "rawMarkdown": "That's great. But are people going to use segmentation models? I tried to use them but since it's segmentation model, I couldn't get the bbox coordinates for COTS.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1669925,
      "author_name": "shigengtian",
      "author_url": "",
      "post_date": "01/31/2022 04:32:09",
      "content": "<p>Thank you for sharing.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1670042,
      "author_name": "xiuqi0",
      "author_url": "",
      "post_date": "01/31/2022 07:04:00",
      "content": "<p>Great! Thanks for sharing👍👍👍</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1670150,
      "author_name": "kfk42kfk",
      "author_url": "",
      "post_date": "01/31/2022 09:00:49",
      "content": "<p>That's great. But are people going to use segmentation models? I tried to use them but since it's segmentation model, I couldn't get the bbox coordinates for COTS.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1669912": "Using the tracking of unique COTS from @hengck23 in https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/299356, I drawed the masks with label-studio of the middle appearance frame for each of the 228 unique COTS. Some of them have a tiny hole inside the mask because of incomplete inpainting. \n\nIn case you want to experiment some adaptation of copy-paste augmentation or any other technique, here is the dataset with the crops of the unique COTS and their associated mask:\nhttps://www.kaggle.com/alexandrecc/cots-masks",
    "1669925": "Thank you for sharing.",
    "1670042": "Great! Thanks for sharing👍👍👍",
    "1670150": "That's great. But are people going to use segmentation models? I tried to use them but since it's segmentation model, I couldn't get the bbox coordinates for COTS."
  },
  "source": "meta"
}