We have started to work on People Count demo application in which we have requirement to count total number of people who are climbing ladder and add into Total count.
We have already done half job to create video streaming using ESP32-CAM and also configured OpenCV with required libraries as well.
But, When we execute provided example at that time we are getting count on every moment of people which is not correct as it will add many counts for single person.
The code reads a video file and processes each frame of the video. It applies the YOLO (You Only Look Once) model to detect people in each frame by drawing bounding boxes around them. It keeps track of the count of detected people and displays it on the video frame. The code also uses non-maximum suppression to eliminate redundant bounding boxes. Finally, it prints the total number of people counted in the video.
Let me know if anyone has any reference or anyone has developed such kind of application before.