I will build computer vision video analytics with yolo detection

Need to detect, track, or count people and objects in video? I build computer vision pipelines with YOLO detection, pose estimation, and multi-object tracking that turn raw footage into clear, usable data.

As lead developer of my graduation research team, I built a violence detection system for CCTV footage that combines detection, pose estimation, and tracking; the paper is currently under peer review at Image and Vision Computing (Elsevier). I bring the same engineering to practical tasks such as people counting, vehicle tracking, and movement analysis.

The starter package processes 1 video of up to 5 minutes and delivers an annotated output video, a CSV file with detections and counts, and the Python script so you can run it on new footage. It uses a standard pre-trained model, which recognizes people, vehicles, and common everyday objects. Training a custom detector on your own labelled images is available as a custom offer.

Please only send footage you have the right to process. Message me with a short description of your video and goal before ordering, and I will confirm the best approach.