Coco Car Damage Detection Dataset

The dataset contains car images with one or more damaged parts. The img/ folder has all 80 images in the dataset. There are three more folders train/, val/ and test/ for training, validation and testing purposes respectively.

Folders

train/:

• Contains 59 images.
• COCO_train_annos.json: Train annotation file for damages where damage is the one and only category.
• COCO_mul_train_annos.json: Train annotation file for parts having damages. There are five categories of parts based on which part the damage has happened. The parts can be namely, headlamp, front_bumper, hood, door, rear_bumper.

val/:

• Contains 11 images.
• COCO_val_annos.json: Validation annotation file for damages where damage is the one and only category.
• COCO_mul_val_annos.json: Validation annotation file for parts having damages. There are five categories of parts based on which part the damage has happened. The parts can be namely, headlamp, front_bumper, hood, door, rear_bumper.

test/:

• Contains 8 images.

Annotation files have the following keys:

1. “annotations”: Contains the bounding box and segmentation array.
2. “categories”: Contains the list of categories in the annotation.
3. “images”: Details of each image used in the annotation.
4. “info”: Creator information
5. “licenses”: License information

Conclusion

Globose Technology Solutions Private Limited is a leader in automotive technology, using the Coco Car Damage Detection Dataset. We are dedicated to innovation and are creating advanced solutions for detecting car damage automatically and accurately. By using information from the Coco dataset, our aim is to make vehicle damage assessment more efficient, which will greatly benefit the automotive industry.