This repository contains minimalistic usage examples of ONNX models for computer vision tasks such as face/object detection. Download and inference codes can be found in their respective folders. You can find a set of faces used for validating clusters in faces, these are randomly taken from LFW (Labeled Faces in the Wild).
Below is an index of existing models in this repository:
| Model Class | Reference | Description | Model/Code Source |
|---|---|---|---|
| ArcFace | Deng et al. | A CNN based model for face recognition which learns discriminative features of faces and produces embeddings for input face images. | InsightFace |
| FaceNet | Schroff et al. | A CNN based model for face recognition that maps face images to a compact Euclidean space where distances directly correspond to a measure of face similarity. | FaceNet Pytorch |
| YOLOv8 | Jocher et al. | A CNN model for real-time object detection that can identify multiple object categories simultaneously. It uses a single network evaluation, making it extremely fast and efficient. This implementation, based on Ultralytics' YOLOv8, offers state-of-the-art performance in speed and accuracy. The model is typically trained on datasets like COCO, capable of detecting 80 common object classes in various scenarios. | YOLOv8, YOLOv8-face, ONNX code |
For more comprehensive list models, visit the original repositry of ONNX Models