React web application for serverless object detection. The inference pipeline is done with ONNX models only and it consists of 4 steps: image preprocessing, object detection, non maximum supression and postprocessing (described in pipeline section). By default digits detection model and 0-9 labels are loaded.

Each pipeline step is done with ONNX models. The complete pipeline during inference is the following:

1. Image preprocessing - resize and pad to match model input size (preprocessing)
2. Object detection - Detect objects with YOLOv8 model (yolo)
3. Non Maximum Supression - Apply NMS to YOLO output (nms)
4. Postprocessing - Apply postprocessing to filtered boxes (postprocessing)

• React - Web application used to test object detection models for real world examples
• ONNX - All processing steps used in pipeline
• ONNX Runtime - Pipeline inference
• OpenCV - Image processing for the preprocessing without ONNX