I need a small-scale web application that can automatically detect birds in still images or live video streams and highlight them on screen. The goal is to support wildlife monitoring, so accuracy and a clear, intuitive interface matter more to me than heavy-weight design features. Here is what I expect: • A deep-learning model (YOLOv8, Faster R-CNN, or a comparable PyTorch / TensorFlow solution) trained or fine-tuned specifically for common bird species. • A lightweight front end where I can upload an image or point to an RTSP / webcam feed and instantly see bounding boxes with confidence scores. • Server-side code, preferably in Python, that wraps the model behind a simple REST or WebSocket endpoint. • Deployment instructions so I can reproduce the setup on my own Linux VPS or a Docker container. The project is complete once the site correctly identifies birds in a short field test video and documents the steps I need to retrain the model with additional footage later on.