Phoenix - Your AI AssistantObject Detection Model — YOLOv8 Custom Object Detection
Delivery in
5 days
- Views 4
Amount of days required to complete work for this Offer as set by the freelancer.
Rating of the Offer as calculated from other buyers' reviews.
Average time for the freelancer to first reply on the workstream after purchase or contact on this Offer.
What you get with this Offer
I will train a custom YOLOv8 object detection model for your specific objects — covering dataset preparation in YOLO format, model training with your labelled images, mAP evaluation per class, NMS threshold tuning, and export in ONNX or TensorRT format for deployment. YOLOv8 is currently the best balance of speed and accuracy available in real-time object detection, making it the right choice for the majority of detection applications — but achieving good performance requires careful handling of class imbalance in bounding box annotations, appropriate anchor configuration, and mosaic augmentation that the default training configuration doesn't always handle correctly for highly specialised object categories.
The model covers dataset preparation and validation, YOLOv8 training with augmentation, mAP@0.5 and mAP@0.5:0.95 evaluation per class, NMS threshold optimisation, model export in ONNX or TensorRT, and an inference script.
The model covers dataset preparation and validation, YOLOv8 training with augmentation, mAP@0.5 and mAP@0.5:0.95 evaluation per class, NMS threshold optimisation, model export in ONNX or TensorRT, and an inference script.
What the Freelancer needs to start the work
Please share your labelled dataset (YOLO format preferred, or COCO/Pascal VOC), your object classes, your deployment target, and your speed vs. accuracy requirements.