I will do object detection and recognition project

You need to provide the dataset (images or video frames) containing the objects you want to detect. Crucially, this data must be annotated with bounding boxes (and masks, if segmentation is required) in a standard format (e.g., YOLO format, COCO format, or Pascal VOC format).

Data annotation is a time-consuming but necessary step. If your data is not yet labeled, I can provide this service as a Gig Extra. Please message me before ordering to discuss the size of your dataset and get a custom quote for annotation, as this affects the project timeline.

I primarily use the most modern and efficient versions, such as YOLO11 and YOLO12, to ensure high accuracy and real-time performance. However, I can also work with older versions (like YOLOv10 or YOLOv8) if your deployment environment requires it.

The trained model weights (e.g., .pt file). The Python source code necessary for running inference (predictions) on new images/videos. A detailed performance report including metrics like mAP (mean Average Precision), Precision, and Recall. Clear documentation on how to run and use the model.