{
  "id": 97201,
  "title": "Understanding Dataset(.csv) ",
  "url": "/competitions/dstl-satellite-imagery-feature-detection/discussion/97201",
  "author_name": "Gaurav Pandey",
  "post_date": "2019-06-26T08:16:39.242000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi,\nI am working on dstl satellite dataset. \nThe .csv file is in the form of multiple polygon system. I have understood that the how the numbers are coordinates and suppose in class building each bracket is the combination of the coordinates for the particular building. However, the images are 3000 x 3000 px but the points are are all less than one(0.007 eg), \nso I wanted so ask how exactly do these numbers signify coordinates in the image.\nAlso how the input image are really big(3000x3000) for a cnn therefore how should i reduce the images while still be able to use the annotations.</p>",
  "messages": [
    {
      "id": 561192,
      "postDate": "2019-06-26T08:16:39.243Z",
      "content": "<p>Hi,\nI am working on dstl satellite dataset. \nThe .csv file is in the form of multiple polygon system. I have understood that the how the numbers are coordinates and suppose in class building each bracket is the combination of the coordinates for the particular building. However, the images are 3000 x 3000 px but the points are are all less than one(0.007 eg), \nso I wanted so ask how exactly do these numbers signify coordinates in the image.\nAlso how the input image are really big(3000x3000) for a cnn therefore how should i reduce the images while still be able to use the annotations.</p>",
      "rawMarkdown": "Hi,\nI am working on dstl satellite dataset. \nThe .csv file is in the form of multiple polygon system. I have understood that the how the numbers are coordinates and suppose in class building each bracket is the combination of the coordinates for the particular building. However, the images are 3000 x 3000 px but the points are are all less than one(0.007 eg), \nso I wanted so ask how exactly do these numbers signify coordinates in the image.\nAlso how the input image are really big(3000x3000) for a cnn therefore how should i reduce the images while still be able to use the annotations."
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "561192": "Hi,\nI am working on dstl satellite dataset. \nThe .csv file is in the form of multiple polygon system. I have understood that the how the numbers are coordinates and suppose in class building each bracket is the combination of the coordinates for the particular building. However, the images are 3000 x 3000 px but the points are are all less than one(0.007 eg), \nso I wanted so ask how exactly do these numbers signify coordinates in the image.\nAlso how the input image are really big(3000x3000) for a cnn therefore how should i reduce the images while still be able to use the annotations."
  }
}