TACO Dataset YOLO Format

The TACO (Trash Annotations in Context) dataset, now available in YOLO (You Only Look Once) format on Kaggle, is a comprehensive resource designed for the detection and classification of litter. Originally provided by ‘Papers with Code,’ this version has been processed for direct usage in YOLO-based object detection models.

TACO includes a diverse range of high-resolution images depicting various types of litter in different contexts and environments. The dataset not only encompasses a broad variety of litter categories commonly found in our surroundings, but it also serves as a valuable asset for training models aimed at environmental cleanup and monitoring purposes.

Each image in this dataset comes with a corresponding annotation file (.txt file) that adheres to the YOLO dataset standard. These annotation files include the coordinates of bounding boxes for the litter present in the image and the respective classes of this litter. Notably, the bounding box annotations are normalized according to the image size, with values ranging from 0 to 1.

The primary goal of this dataset is to support the development of robust and accurate object detection models for litter identification and classification. By doing so, it aims to help create effective solutions for environmental problems, such as pollution and littering. Additionally, this can contribute to the development of automated cleanup systems.

Although the dataset is not divided into separate training, validation, or testing subsets, users are encouraged to create these divisions according to their model development requirements.

Please ensure you follow the terms and conditions specified by the original dataset providers when using this dataset. Additionally, if you find this dataset beneficial for your research or project, consider citing the original source to acknowledge the creators’ efforts.