Upload a photo of your tech gear and watch a custom-trained object detection model identify what's in the frame — in real time.
Capabilities
YOLOv8 model trained on personal photos of real tech equipment — laptops, keyboards, mice, ethernet cables, power bricks, and more.
Server-side inference via FastAPI and ONNX Runtime. Upload an image and get bounding boxes and confidence scores in under a second.
Every detection is logged. The dashboard shows object counts, confidence trends, and your most frequently detected items over time.
Animated canvas overlay, detection result cards, confidence bars, and stat counters — designed to feel like a real AI lab interface.
Pipeline
Drop a photo or pick a file from your device
Image is posted to the FastAPI backend
YOLOv8 runs inference on the server GPU
Detections with labels, confidence, and boxes
Canvas draws boxes, panel shows results