I’m putting together a computer-vision model that can spot household objects in real-time video or still images. The goal is broad coverage: I need the system to recognise kitchen appliances, furniture and electronics no matter where they appear—kitchen, living room or bedroom, plus spaces that rarely get attention such as the balcony, study, drawing room and dining room. You are free to work in the framework you know best (YOLOv8, Detectron2, TensorFlow, PyTorch, OpenCV, etc.) as long as the final solution runs on a consumer-grade GPU and can be called from Python. Pre-trained weights are fine if they shorten development time, but the finished model must be fine-tuned so that it reliably detects each category I listed above. Deliverables • A trained model able to detect kitchen appliances, furniture and electronics across all listed rooms • Inference script (Python) with clear instructions on how to run it on new images or live video • A short report outlining dataset sources, augmentation steps and achieved accuracy • Simple guide showing how to add new classes or rooms later without retraining from scratch If you have experience building custom object-detection datasets or optimising models for edge deployment, let me know in your bid.