{
  "id": 240682,
  "title": "DIY: resized images **with** bounding boxes",
  "url": "/competitions/siim-covid19-detection/discussion/240682",
  "author_name": "Konrad Banachewicz",
  "post_date": "2021-05-21T01:16:31.286000",
  "votes": 11,
  "comment_count": 0,
  "views": 0,
  "content": "<p>I decided to start looking at image level data, so resizing the images and the bounding boxes became kind of crucial. The kernel I used can be found here:</p>\n<p><a href=\"https://www.kaggle.com/konradb/diy-rescaled-images-with-bboxes\" target=\"_blank\">https://www.kaggle.com/konradb/diy-rescaled-images-with-bboxes</a></p>\n<p>and the resulting datasets below:</p>\n<p>128x128: <a href=\"https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-128x128\" target=\"_blank\">https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-128x128</a><br>\n256x256: <a href=\"https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-256x256\" target=\"_blank\">https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-256x256</a><br>\n512x512: in progress</p>\n<p>The accompanying tabular data has two columns: image id and resized label (in the format matching the original).  This should make life a little easier in the experimentation phase.</p>",
  "messages": [
    {
      "id": 1316878,
      "postDate": "2021-05-21T01:16:31.287Z",
      "content": "<p>I decided to start looking at image level data, so resizing the images and the bounding boxes became kind of crucial. The kernel I used can be found here:</p>\n<p><a href=\"https://www.kaggle.com/konradb/diy-rescaled-images-with-bboxes\" target=\"_blank\">https://www.kaggle.com/konradb/diy-rescaled-images-with-bboxes</a></p>\n<p>and the resulting datasets below:</p>\n<p>128x128: <a href=\"https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-128x128\" target=\"_blank\">https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-128x128</a><br>\n256x256: <a href=\"https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-256x256\" target=\"_blank\">https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-256x256</a><br>\n512x512: in progress</p>\n<p>The accompanying tabular data has two columns: image id and resized label (in the format matching the original).  This should make life a little easier in the experimentation phase.</p>",
      "rawMarkdown": "I decided to start looking at image level data, so resizing the images and the bounding boxes became kind of crucial. The kernel I used can be found here:\n\nhttps://www.kaggle.com/konradb/diy-rescaled-images-with-bboxes\n\nand the resulting datasets below:\n\n128x128: https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-128x128\n256x256: https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-256x256\n512x512: in progress\n\nThe accompanying tabular data has two columns: image id and resized label (in the format matching the original).  This should make life a little easier in the experimentation phase.",
      "votes": 11
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1316878": "I decided to start looking at image level data, so resizing the images and the bounding boxes became kind of crucial. The kernel I used can be found here:\n\nhttps://www.kaggle.com/konradb/diy-rescaled-images-with-bboxes\n\nand the resulting datasets below:\n\n128x128: https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-128x128\n256x256: https://www.kaggle.com/konradb/resized-images-with-bounding-boxes-256x256\n512x512: in progress\n\nThe accompanying tabular data has two columns: image id and resized label (in the format matching the original).  This should make life a little easier in the experimentation phase."
  }
}