{
  "fetched_at": "2026-10-19T09:00:16Z",
  "source_url": "https://www.kaggle.com/competitions/cars-object-detection/data",
  "pages": [
    {
      "name": "data-description",
      "content": "This dataset is structured for an object detection task. The dataset includes three files. The annotations are in YOLO (You Only Look Once) format.\n\n### Files Provided\n\n1. **train.zip**\n   - Contains two directories:\n     - `train/images`: This directory contains the JPEG images for training.\n     - `train/labels`: This directory contains corresponding YOLO format label files for the training images. Each label file shares the same name as its corresponding image and includes annotations.\n\n2. **test.zip**\n   - Contains one directory:\n     - `test/images`: This directory contains the JPEG images for testing. These images are used to evaluate the final model predictions in the competition.\n\n3. **sample_submission.csv**\n   - A sample submission file in the correct format that participants should use as a template for submitting their predictions.\n\n### Data Format in Labels\n\nEach label file in the `train/labels` directory contains annotations for the corresponding image in the `train/images` directory. The format for each annotation within the label files is as follows:\n- **class_id**: An integer representing the category of the object. The problem has only one class, so the value is always 0.\n- **center_x**: The x-coordinate of the center of the object, normalized by the image width (0 to 1).\n- **center_y**: The y-coordinate of the center of the object, normalized by the image height (0 to 1).\n- **width**: The width of the object, normalized by the image width (0 to 1).\n- **height**: The height of the object, normalized by the image height (0 to 1).\n\nHere’s an example of what the content of a label file might look like:\n```\n0 0.512 0.469 0.230 0.320\n0 0.230 0.350 0.120 0.156\n```\nIn this example, each line corresponds to a different object within the image. The first number is the class ID, followed by the normalized center coordinates, width, and height of the object's bounding box.\n\n### Explanation of `sample_submission.csv` Format\n\nThe `sample_submission.csv` file provides a format guide for submitting the predictions of the object detection model.\n\n**File Structure**\n\nThe `sample_submission.csv` contains two columns:\n\n1. **id**: This is the identifier corresponding to the filename of an image from the `test/images` directory in the `test.zip` file. The ID should be the filename without the '.jpg' extension.\n   \n2. **predictions**: This column contains the predicted annotations for each image, formatted as *List*. Each prediction in the *List* consists of several components related to the detected objects in the image, formatted as follows:\n   - **class_id**: The predicted class identifier for the object.\n   - **confidence**: The confidence score of the detection.\n   - **center_x**: The normalized x-coordinate of the center of the bounding box.\n   - **center_y**: The normalized y-coordinate of the center of the bounding box.\n   - **width**: The normalized width of the bounding box.\n   - **height**: The normalized height of the bounding box.\n\nHere is the format for each prediction entry in the `predictions` column:\n```\n[class_id, confidence, center_x, center_y, width, height, class_id, confidence, center_x, center_y, width, height]\n```\n\n**Example Entry**\n\nSuppose an image with ID '3ee1ee02-GOPR6568' from the test set has two objects detected. And image with ID 'aece3643-GOPR6651' from the test set has no objects detected. The `sample_submission.csv` entry might look like this:\n\n```csv\nid, predictions\n3ee1ee02-GOPR6568,\"[0, 0.78, 0.44732, 0.91217, 0.01795, 0.02063, 0, 1.0, 0.81899, 0.78349, 0.02522, 0.01389]\"\naece3643-GOPR6651,[]\n```\nIn this example:\n- The first object detected is of class 0 with a confidence of 0.78. Its bounding box center is at (0.44732, 0.91217) with a width of 0.01795 and a height of 0.02063.\n- The second object is of class 0 with a confidence of 1.0, with the center of its bounding box at (0.81899, 0.78349), and dimensions of 0.02522 in width and 0.01389 in height.\n\n**Important Notes**\n\n- class_id is always 0\n- Ensure that `center_x`, `center_y`, `width`, and `height` are normalized values relative to the dimensions of the respective image. \n- Confidence scores should be between 0 and 1, where 1 indicates the highest confidence.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "In this competition, participants are provided with a large set of images along with their corresponding annotations in a training set. The test set contains images without annotations. The task is to predict the bounding boxes for a single object class (car) in these images as accurately as possible.\n\nThe images cover a variety of scenes of parking spaces, with cars of interest that vary widely in size, shape, and clarity.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "## Evaluation\n\n### Metric\nSubmissions are evaluated using the mean Average Precision (mAP) at an Intersection over Union (IoU) threshold of 50%. Only predictions with a confidence level greater than 50% are considered.\n\nhttps://www.kaggle.com/code/its7171/map-understanding-with-code-and-its-tips#The-lower-the-confidence-threshold-is-better",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "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 - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "Welcome to the Object Detection Challenge! This competition is designed to advance the field of computer vision by challenging participants to develop models that can accurately detect objects in images. The focus is on evaluating the model's ability to predict bounding boxes around objects using mean Average Precision (mAP) with an Intersection over Union (IoU) threshold of 50%.\n**Goal:** \n\nThe goal of this competition is to create a model that can accurately identify and localize objects in various images. Participants are expected to predict bounding boxes around the objects with a confidence score greater than 50%. The challenge will help in understanding the effectiveness of different object detection algorithms in varying conditions and their applicability in real-world scenarios.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}