{
  "fetched_at": "2026-10-19T12:32:12Z",
  "source_url": "https://www.kaggle.com/competitions/driver-gaze-zone-classification/data",
  "pages": [
    {
      "name": "data-description",
      "content": "The dataset consists of 2 folders: 'train' and 'test'. The folder 'train' is a labelled dataset and consists of 5 sub-folders: Forward, Centerstack, Right_wing_mirror, Rearview_mirror, and Left_wing_mirror. Each of these sub-folders contains 500 images each corresponding to each class. All images are of the ‘.jpg’ format.\n\nThe ‘test’ folder consists of the test dataset images. There are a total of 509 images in this folder (all ‘.jpg’ files). These images are unlabelled i.e., the labels are not provided. The objective of this challenge is to predict the gaze zone class of all test images. The filenames of these test images are given as 1.jpg, 2.jpg, 3.jpg, …, 509.jpg.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "This assignment is on image classification problem using deep convolutional neural networks. The dataset consists of images of driver gaze towards 5 different zones: *Forward*, *Centerstack*, *Right wing mirror*, *Rearview mirror*, and *Left wing mirror.* Your objective in this challenge is to develop a image classification model that can classify from a given image to it’s corresponding class i.e., driver gaze zone. \n\n\n## Acknowledgements\nThis dataset is adapted from the following research article:\nShreya Ghosh, Abhinav Dhall, Garima Sharma, Sarthak Gupta and Nicu Sebe. Speak2Label: Using Domain Knowledge for Creating a Large Scale Driver Gaze Zone Estimation Dataset. ICCVW 2021. https://arxiv.org/abs/2004.05973\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<p>The evaluation metric for this challenge is <a href=\"https://www.kaggle.com/wiki/MeanFScore\"> Mean F1-Score</a>. The F1 score, commonly used in information retrieval, measures accuracy using the statistics precision p and recall r. Precision is the ratio of true positives (tp) to all predicted positives (tp + fp). Recall is the ratio of true positives to all actual positives (tp + fn). The F1 score is given by:</p>\n<p>\\[ F1 = 2\\frac{p \\cdot r}{p+r}\\ \\ \\mathrm{where}\\ \\ p = \\frac{tp}{tp+fp},\\ \\ r = \\frac{tp}{tp+fn} \\]</p>\n<p>The F1 metric weights recall and precision equally, and a good retrieval algorithm will maximize both precision and recall simultaneously. Thus, moderately good performance on both will be favored over extremely good performance on one and poor performance on the other.</p>\n<p>50% of the test images will be used for public leaderboard performance calculation and the remaining 50% will be used for private leaderboard. Public scores will be shown on the public leaderboard, while only the private scores will be used for final evaluation. The private leaderboard score is not shown until the end of the competititon. You can hand-select any 2 of the eligible submissions for private leaderboard score, or will otherwise default to the best public scoring submissions.</p>\n\n\n<h2>Submission Format</h2>\n<p> Submission files should be in .csv format and contain two columns: filename and class. ‘filename’ should be in string format and will contain the image filename (1.jpg, 2. jpg, etc.). The ‘class’ will also be in string format and contain the corresponding class name.. It can take any one of the 5 values: Forward, Centerstack, Right_wing_mirror, Rearview_mirror, and Left_wing_mirror. Since there are 509 images in the test dataset, your submission should have 509 rows (excluding the header), each with 2 columns. A sample submission file is also provided in the dataset . A sample submission file is also provided in the dataset.  </p>\n<p>The file should contain a header and have the following format:</p>\n<pre>filename,class<br>1.jpg,Forward<br></pre>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}