price-tag-detection

# Price tag detection dataset This dataset was used to train the price tag object detection model running in production for the Open Prices project. Images were collected from the Open Prices database and labeled manually. Just like the original images, the images in this dataset are licensed under the Creative Commons Attribution Share Alike license (CC-BY-SA 4.0). Images were resized such as the longest size of the image is not greater than 1024px, while keeping the original aspect ratio. ## Fields - `image_id`: Unique identifier for the image, generated from full image path. - `image`: Image data. - `width`: Image in pixels. - `height`: Image height in pixels. - `meta`: Additional metadata. - `image_url`: original URL of the image. - `objects`: Object detection annotations. - `bbox`: List of bounding boxes in the format (y_min, x_min, y_max, x_max). Coordinates are normalized between 0 and 1, using the top-left corner as the origin. - `category_id`: List of category IDs. - `category_name`: List of category names. ## Versions - `1.2`: Removed incorrect bounding box and added a few more samples. - `1.1`: Addition of new samples in `train` and `val` splits, for which the previous model performed incorrect detections. See [this report](https://github.com/openfoodfacts/openfoodfacts-ai/blob/develop/docs/reports/2026-01-price-tag-detection.md) for more details. - `1.0`: First version released.