YOLO-Face-Person-Detector

Model Overview

This model is a fine-tuned YOLOv8x detector specialized in two classes: Face and Person. It is hosted as a managed, OpenAI-compatible API on gigarouter, requiring no local installation or model loading.

Key Strengths

• Trained on a proprietary dataset of approximately 150,000 images, providing high accuracy and robustness across diverse scenarios.
• Built on the YOLOv8x architecture, offering strong detection capacity for real-time applications.
• Optimized for simultaneous face and person detection in a single forward pass.

Best For

• Applications requiring joint face and person detection, such as surveillance, crowd analysis, photo organization, and content moderation.
• Scenarios where a single model can replace separate face and person detectors, reducing latency and system complexity.

Benchmark Results

Specific benchmark numbers are not provided in the model card. The model is described as achieving high accuracy on its proprietary training set, but no public dataset metrics (e.g., COCO, WIDER Face) are cited.

Sample Output

License

This model inherits the AGPL-3.0 License from Ultralytics YOLOv8. Refer to the Ultralytics Licensing page for commercial usage details.

Citation

If you use this model in research, please cite:

• Kuprashevich et al., "MiVOLO: Multi-input Transformer for Age and Gender Estimation", 2023.
• Kuprashevich et al., "Beyond Specialization: Assessing the Capabilities of MLLMs in Age and Gender Estimation", 2024.
• Tolstykh et al., "CerberusDet: Unified Multi-Dataset Object Detection", 2024.