COCO Dataset and YOLO Object Detection

[maryam-mohebbi]

Implement YOLO object detection using the COCO dataset.

• Familiarize with and preprocess the COCO dataset for object detection.
• Implement and train the YOLO (You Only Look Once) model for object detection.
• Test and optimize the YOLO model for accuracy and efficiency on the COCO dataset.

[maryam-mohebbi]

Related links: https://arxiv.org/pdf/1508.06576.pdf https://arxiv.org/pdf/1605.04603.pdf https://www.semanticscholar.org/reader/f8e79ac0ea341056ef20f2616628b3e964764cfd https://paperswithcode.com/dataset/coco#:~:text=The%20MS%20COCO%20,41K%29%20sets

[maryam-mohebbi]

Maybe it is worth looking at this dataset: https://www.cvlibs.net/datasets/kitti/index.php

• Left color images of object data set (12 GB): This contains the actual images in which you will detect objects.

• Label files (5 MB): These files contain the annotations for the objects in the images, which are used to train and test object detection models. It will have the 2D bounding box information for each object.