The project centers on building a customized YOLOv8 model that detects vehicles in images or video, then documenting the entire workflow in a concise, publication-ready scientific report. I have already earmarked ₹1,500 for the coding component and ₹1,000 for the writing, and everything has to come together within one week. What the coding phase entails • Train and fine-tune YOLOv8 in Python (Ultralytics implementation) for multi-class vehicle detection. • Deliver a clean, reproducible codebase with clear folder structure, requirements.txt, and an inference script that runs on CPU or GPU. • Include a short README that explains environment setup, training commands, and how to test the detector on new footage. What the writing phase covers • Compose a well-formatted scientific report (≈8–10 pages) in MS Word or LaTeX—your choice—summarizing data preparation, model architecture, training regimen, evaluation metrics, results, and future work. • Integrate plots, confusion matrices, sample detections, and relevant citations. • Ensure the language reads like a conference paper: abstract, introduction, methodology, experiments, results, conclusion, references. Acceptance criteria 1. YOLOv8 model reaches reasonable precision and recall on the test set and runs inference in real time (≥15 FPS on 720p video with GPU). 2. Code executes end-to-end on my machine without manual fixes. 3. Report is grammatically polished, logically structured, and ready for direct submission. I’m available for quick feedback throughout the week to keep things on schedule. Let’s get started right away so we can hit the deadline together.