README.md

---
license: agpl-3.0
pipeline_tag: object-detection
base_model:
  - Ultralytics/YOLOv8
tags:
- ultralytics
- object-detection
- yolov8
- quantized
- int8
---

<div align="center">
  <p>
    <a href="https://www.ultralytics.com/events/yolovision/?utm_source=nxp&utm_medium=referral&utm_campaign=partner" target="_blank">
      <img width="100%" src="https://cdn.prod.website-files.com/680a070c3b99253410dd3dcf/68e4ed994ba1820276d637b8_Ultralytics%20YOLOv8.svg" alt="YOLO Vision banner"></a>
  </p>

<div>
    <a href="https://huggingface.co/nxp/YOLOv8/blob/main/LICENSE.txt"><img src="https://img.shields.io/badge/License-AGPL%203.0-blue" alt="License AGPL-3.0"></a>
    <a href="https://www.ultralytics.com/"><img alt="Static Badge" src="https://img.shields.io/badge/Ultralytics-blue?logo=ultralytics&label=Origin&link=https%3A%2F%2Fwww.ultralytics.com%2F"></a>
</div>
</div>

# Overview

[Ultralytics](https://www.ultralytics.com/?utm_source=nxp&utm_medium=referral&utm_campaign=partner) released [Ultralytics YOLOv8](https://docs.ultralytics.com/models/yolov8/?utm_source=nxp&utm_medium=referral&utm_campaign=partner) on January 10, 2023, delivering state‑of‑the‑art accuracy and speed. Building on the progress of earlier Ultralytics YOLO versions, it introduces improved features and optimizations that make it a strong choice for a wide range of object detection, segmentation and pose estimation tasks across many applications.

## Key Features of Ultralytics YOLOv8

* Advanced Backbone and Neck Architectures: YOLOv8 incorporates modern backbone and neck designs that enhance feature extraction and overall detection performance.
* Anchor‑Free Split Ultralytics Head: By using an anchor‑free split head, YOLOv8 achieves higher accuracy and more efficient detection compared to traditional anchor‑based methods.
* Balanced Accuracy–Speed Optimization: YOLOv8 is engineered to deliver strong accuracy while maintaining real‑time speed, making it well‑suited for a wide range of real‑time detection applications.
* Multiple Pretrained Model Options: A variety of pretrained models are available, allowing users to choose the one that best matches their task requirements and performance needs.

<div align="center">
  <p>
    <a href="https://docs.ultralytics.com/models/yolov8/?utm_source=nxp&utm_medium=referral&utm_campaign=partner" target="_blank">
      <img width="100%" src="https://cdn.jsdelivr.net/gh/ultralytics/assets@main/docs/yolov8-comparison-plots.avif" alt="YOLOv8 performance comparison charts"></a>
  </p>
</div>

## Model Description

This is a repository that contains a set of quantized and compiled versions of [Ultralytics YOLOv8](https://docs.ultralytics.com/models/yolov8/) models optimized for Ara240 DNPU.

- **Base Model:** [Ultralytics/YOLOv8](https://github.com/ultralytics/ultralytics)
- **Original Model Authors:** Ultralytics
- **Original License:** AGPL-3.0
- **Modified by:** NXP
          
### Modifications

This model is a derivative work with the following changes from the original:

- **Quantization:** INT8 calibrated using [COCO val2017](https://cocodataset.org/)
- **Compilation:** Compiled for Ara240 DNPU
- **Format:** Converted to DVM format for NPU deployment

Original model available at: [Ultralytics YOLOv8](https://docs.ultralytics.com/models/yolov8/).

## Performance Summary

### Object Detection

| Model   | size<br><sup>(pixels) | FP32 mAP<sup>val<br>50-95 | INT8 mAP<sup>val<br>50-95 | Latency <br><sup>Ara240<br>(ms) | Performance<br><sup>Ara240<br>(inferences/s) | params<br><sup>(M) |
| ------- | --------------------- | ------------------------- | ------------------------- | ------------------------------- | -------------------------------------------- | ------------------ |
| YOLOv8n | 640                   | 37.3                      | 35.5                      | 3.15                            | 316.99                                       | 3.2                |
| YOLOv8s | 640                   | 44.9                      | 43.4                      | 6.72                            | 148.72                                       | 11.2               |
| YOLOv8m | 640                   | 50.2                      | 48.7                      | 17.60                           | 56.80                                        | 25.9               |
| YOLOv8l | 640                   | 52.9                      | 51.4                      | 35.96                           | 27.80                                        | 43.7               |
| YOLOv8x | 640                   | 53.9                      | 49.7                      | 56.02                           | 17.84                                        | 68.2               |

- **mAP<sup>val</sup>** values are for single-model single-scale on [COCO val2017](https://cocodataset.org/) dataset.

### Segmentation

| Model       | size<br><sup>(pixels) | FP32 mAP<sup>mask<br>50-95 | INT8 mAP<sup>mask<br>50-95 | Latency <br><sup>Ara240<br>(ms) | Performance<br><sup>Ara240<br>(inferences/s) | params<br><sup>(M) |
| ----------- | --------------------- | -------------------------- | -------------------------- | ------------------------------- | -------------------------------------------- | ------------------ |
| YOLOv8n-seg | 640                   | 30.5                       | 29.18                      | 4.13                            | 241.88                                       | 3.4                |
| YOLOv8s-seg | 640                   | 36.8                       | 36.10                      | 10.14                           | 98.61                                        | 11.8               |
| YOLOv8m-seg | 640                   | 40.8                       | 39.71                      | 25.07                           | 39.88                                        | 27.3               |
| YOLOv8l-seg | 640                   | 42.6                       | 39.15                      | 49.87                           | 20.04                                        | 46                 |
| YOLOv8x-seg | 640                   | 43.4                       | 39.6                       | 78.91                           | 12.67                                        | 71.8               |

- **mAP<sup>mask</sup>** values are for single-model single-scale on [COCO val2017](https://cocodataset.org/) dataset.

### Pose

| Model        | size<br><sup>(pixels) | FP32 mAP<sup>pose<br>50-95 | INT8 mAP<sup>pose<br>50-95 | Latency <br><sup>Ara240<br>(ms) | Performance<br><sup>Ara240<br>(inferences/s) | params<br><sup>(M) |
| ------------ | --------------------- | -------------------------- | -------------------------- | ------------------------------- | -------------------------------------------- | ------------------ |
| YOLOv8n-pose | 640                   | 50.4                       | 46.75                      | 4.08                            | 245.01                                       | 3.3                |
| YOLOv8s-pose | 640                   | 60                         | 53.66                      | 8.07                            | 123.88                                       | 11.6               |
| YOLOv8m-pose | 640                   | 65                         | 58.84                      | 19.02                           | 52.56                                        | 26.4               |
| YOLOv8l-pose | 640                   | 67.6                       | 62.84                      | 36.85                           | 27.13                                        | 44.4               |
| YOLOv8x-pose | 640                   | 69.2                       | 64.76                      | 59.48                           | 16.8                                         | 69.4               |

- **mAP<sup>pose</sup>** values are for single-model single-scale on [COCO val2017](https://cocodataset.org/) dataset.

## License

Ultralytics offers two licensing options to accommodate diverse use cases:

- **AGPL-3.0 License**: This [OSI-approved](https://opensource.org/license) open-source license is ideal for students and enthusiasts, promoting open collaboration and knowledge sharing. See the [LICENSE](https://www.ultralytics.com/license/?utm_source=nxp&utm_medium=referral&utm_campaign=partner) file for more details.
- **Enterprise License**: Designed for commercial use, this license permits seamless integration of Ultralytics software and AI models into commercial goods and services, bypassing the open-source requirements of AGPL-3.0. If your scenario involves embedding our solutions into a commercial offering, reach out through [Ultralytics Licensing](https://www.ultralytics.com/license/?utm_source=nxp&utm_medium=referral&utm_campaign=partner).