Awesome YOLOv8 Models

Easy-to-use finetuned YOLOv8 models

Demo • Github

TABLE OF CONTENTS

1. About the Project
2. Installation
3. Usage
4. Classification Models
5. Detection Models
6. Segmentation Models
7. Contributing
8. License

About the Project

This is a collection of YOLOv8 models finetuned for classification/detection/segmentation tasks on datasets from various domains as Medicine/Insurance/Sports/Gaming.

Ultralytics YOLOv8, developed by Ultralytics, is a cutting-edge, state-of-the-art (SOTA) model that builds upon the success of previous YOLO versions and introduces new features and improvements to further boost performance and flexibility. YOLOv8 is designed to be fast, accurate, and easy to use, making it an excellent choice for a wide range of object detection, image segmentation and image classification tasks.

_Source: github

Installation

To use listed models, install ultralyticsplus:

pip install ultralyticsplus

Usage

from ultralyticsplus import YOLO, render_result

# load model
model = YOLO(DESIRED_MODEL_ID)

# set image
image = 'image.png'
    
# perform inference
results = model(image)

# parse results
result = results[0]
boxes = result.boxes.xyxy # x1, y1, x2, y2
scores = result.boxes.conf
categories = result.boxes.cls
scores = result.probs # for classification models
masks = result.masks # for segmentation models
    
# show results on image
render = render_result(model=model, image=image, result=result)
render.show()

Classification Models

top1 acc.	top5 acc.	model type	model id	dataset page
0.678	1.000	yolov8n-cls	keremberke/yolov8n-shoe-classification	dataset
0.687	1.000	yolov8s-cls	keremberke/yolov8s-shoe-classification	dataset
0.795	1.000	yolov8m-cls	keremberke/yolov8m-shoe-classification	dataset

top1 acc.	top5 acc.	model type	model id	dataset page
0.943	1.000	yolov8n-cls	keremberke/yolov8n-chest-xray-classification	dataset
0.942	1.000	yolov8s-cls	keremberke/yolov8s-chest-xray-classification	dataset
0.955	1.000	yolov8m-cls	keremberke/yolov8m-chest-xray-classification	dataset

Detection Models

box mAP@0.5	model type	model id	dataset page
0.937	yolov8n	keremberke/yolov8n-valorant-detection	dataset
0.971	yolov8s	keremberke/yolov8s-valorant-detection	dataset
0.965	yolov8m	keremberke/yolov8m-valorant-detection	dataset

box mAP@0.5	model type	model id	dataset page
0.838	yolov8n	keremberke/yolov8n-forklift-detection	dataset
0.851	yolov8s	keremberke/yolov8s-forklift-detection	dataset
0.846	yolov8m	keremberke/yolov8m-forklift-detection	dataset

box mAP@0.5	model type	model id	dataset page
0.844	yolov8n	keremberke/yolov8n-csgo-player-detection	dataset
0.886	yolov8s	keremberke/yolov8s-csgo-player-detection	dataset
0.892	yolov8m	keremberke/yolov8m-csgo-player-detection	dataset

box mAP@0.5	model type	model id	dataset page
0.893	yolov8n	keremberke/yolov8n-blood-cell-detection	dataset
0.917	yolov8s	keremberke/yolov8s-blood-cell-detection	dataset
0.927	yolov8m	keremberke/yolov8m-blood-cell-detection	dataset

box mAP@0.5	model type	model id	dataset page
0.995	yolov8n	keremberke/yolov8n-plane-detection	dataset
0.995	yolov8s	keremberke/yolov8s-plane-detection	dataset
0.995	yolov8m	keremberke/yolov8m-plane-detection	dataset

box mAP@0.5	model type	model id	dataset page
0.209	yolov8n	keremberke/yolov8n-nlf-head-detection	dataset
0.279	yolov8s	keremberke/yolov8s-nlf-head-detection	dataset
0.287	yolov8m	keremberke/yolov8m-nlf-head-detection	dataset

box mAP@0.5	model type	model id	dataset page
0.836	yolov8n	keremberke/yolov8n-hard-hat-detection	dataset
0.834	yolov8s	keremberke/yolov8s-hard-hat-detection	dataset
0.811	yolov8m	keremberke/yolov8m-hard-hat-detection	dataset

box mAP@0.5	model type	model id	dataset page
0.967	yolov8n	keremberke/yolov8n-table-extraction	dataset
0.984	yolov8s	keremberke/yolov8s-table-extraction	dataset
0.952	yolov8m	keremberke/yolov8m-table-extraction	dataset

Segmentation Models

mask mAP@0.5	model type	model id	dataset page
0.491	yolov8n-seg	keremberke/yolov8n-pcb-defect-segmentation	dataset
0.517	yolov8s-seg	keremberke/yolov8s-pcb-defect-segmentation	dataset
0.557	yolov8m-seg	keremberke/yolov8m-pcb-defect-segmentation	dataset

mask mAP@0.5	model type	model id	dataset page
0.628	yolov8n-seg	keremberke/yolov8n-building-segmentation	dataset
0.651	yolov8s-seg	keremberke/yolov8s-building-segmentation	dataset
0.613	yolov8m-seg	keremberke/yolov8m-building-segmentation	dataset

mask mAP@0.5	model type	model id	dataset page
0.995	yolov8n-seg	keremberke/yolov8n-pothole-segmentation	dataset
0.928	yolov8s-seg	keremberke/yolov8s-pothole-segmentation	dataset
0.895	yolov8m-seg	keremberke/yolov8m-pothole-segmentation	dataset

Contributing

To contribute to Awesome-YOLOv8-Models, follow these steps:

1. Train a YOLOv8 model with ultralytics package | tutorial
2. Push your model to hub with ultralyticsplus package | package readme
3. Open a PR or Discussion post in this repo with your hub id.

License

This project is licensed under MIT license. See LICENSE for more information.

Give a ⭐️ if this project helped you!