{
  "fetched_at": "2026-10-25T00:18:33Z",
  "source_url": "https://www.kaggle.com/competitions/dl4cv1/data",
  "pages": [
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "data-description",
      "content": "<h3>This is a simple Object Localisation task to detect the bounding boxes around an object in an image</h3>\n<p>Your task is to regress over the four parameters and output the 4 co-ordinates of the bounding box</p>\n<b>The data of the images to be used for training and testing can be found here: </b> \n<href> https://drive.google.com/open?id=1FbyOAkNoR0mZsaj6wj93UYtmL5zOSCq9</href>\n\n<b>The evaluation metric used for this assignment is DICE</b>\n<p>Simply put, the Dice Coefficient is 2 * the Area of Overlap divided by the total number of pixels in both images the code for which is :</p>\n<p> </p>\n<b>def dice_coef(y_true, y_pred, smooth=1): </b>\n<p> &nbsp intersection = K.sum(y_true * y_pred, axis=[1,2,3])  </p>\n<p> &nbsp union = K.sum(y_true, axis=[1,2,3]) + K.sum(y_pred, axis=[1,2,3])  </p>\n<p> &nbsp dice = K.mean((2. * intersection + smooth)/(union + smooth), axis=0)  </p>\n<p> &nbsp return dice </p>\n\n<p>Please note that you will need to convert your bounding boxes into a run-length encoding for submission</p>\n<b> What is Run length encoding? </b>\n<p> In order to reduce the submission file size, our metric uses run-length encoding on the pixel values. Instead of submitting an exhaustive list of indices for your segmentation, you will submit pairs of values that contain a start position and a run length. E.g. '1 3' implies starting at pixel 1 and running a total of 3 pixels (1,2,3).\n\nThe competition format requires a space delimited list of pairs. For example, '1 3 10 5' implies pixels 1,2,3,10,11,12,13,14 are to be included in the mask. The pixels are one-indexed and numbered from top to bottom, then left to right: 1 is pixel (1,1), 2 is pixel (2,1), etc. </p>\n\n<b> The code for converting a dataframe with bounding box co-ordinates into Run-length encoding is present in the notebooks section. Please convert into the submission format before submitting. </b>\n\n\n\n\n\n\n<h2>File descriptions</h2>\n<ul>\n<li><strong>train_final.csv</strong> - the training set</li>\n<li><strong>testing_final3.csv</strong> - the test set</li>\n<li><strong>sampleSubmission.csv</strong> - a sample submission file in the correct format</li>\n</ul>\n<h2>Data fields</h2>\n<ul>\n<li><strong>ids</strong> - Image names</li>\n<li><strong>x1</strong> - Starting x co-ordinate of box</li>\n<li><strong>x2</strong> - Ending x co-ordinate of box</li>\n<li><strong>y2</strong> - Starting y co-ordinate of box</li>\n<li><strong>y2</strong> - Ending y co-ordinate of box</li>\n<li><strong>rle</strong> - Run length encoding of the bounding box</li>\n</ul>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<h3> Assignment for the course Deep Learning for Computer Vision as part of the CTE (Center for Technical Education) initiative </h3>\n\n<p> Please check the data section for more details </p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<b>The evaluation metric used for this assignment is DICE</b>\n<p>Simply put, the Dice Coefficient is 2 * the Area of Overlap divided by the total number of pixels in both images the code for which is :</p>\n<p> </p>\n<b>def dice_coef(y_true, y_pred, smooth=1): </b>\n<p> &nbsp intersection = K.sum(y_true * y_pred, axis=[1,2,3])  </p>\n<p> &nbsp union = K.sum(y_true, axis=[1,2,3]) + K.sum(y_pred, axis=[1,2,3])  </p>\n<p> &nbsp dice = K.mean((2. * intersection + smooth)/(union + smooth), axis=0)  </p>\n<p> &nbsp return dice </p>\n\n<p>Please note that you will need to convert your bounding boxes into a run-length encoding for submission</p>\n<b> What is Run length encoding? </b>\n<p> In order to reduce the submission file size, our metric uses run-length encoding on the pixel values. Instead of submitting an exhaustive list of indices for your segmentation, you will submit pairs of values that contain a start position and a run length. E.g. '1 3' implies starting at pixel 1 and running a total of 3 pixels (1,2,3).\n\nThe competition format requires a space delimited list of pairs. For example, '1 3 10 5' implies pixels 1,2,3,10,11,12,13,14 are to be included in the mask. The pixels are one-indexed and numbered from top to bottom, then left to right: 1 is pixel (1,1), 2 is pixel (2,1), etc. </p>\n\n<b> The code for converting a dataframe with bounding box co-ordinates into Run-length encoding is present in the notebooks section. Please convert into the submission format before submitting. </b>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}