I finished this object detection project when I am a master student in NUAA.The task is to find whether the driver is in the car.
l follow the blog by Guanghan Ning (http://guanghan.info/blog/en/my-works/train-yolo/). And the blog is about start training YOLO with our own data and object classes.
In order to finish this project,I divide this task in three steps.
First, I set a dataset about two classes which are people sit in the chair and seat back.
Second, Using the YOLO real-time object detection method and the idea in the blog I proposed above, I train an model in my own dataset.
Third, I test the model by using the test image and video. In the end ,I get a well enough result in my project.
Tips: If you want to use my dataset,please tell me. Thanks!