Using a video source (webcam or loaded video), generate a visual heat map or motion map, count the number of people present, and analyze the results. Create an output video and save a specific snapshot of it to send to the cloud.
Gain insight into the following solutions:
Computer vision applications for IoT
Inference to analyze datasets
Retail market IoT
Learn to build and run an application with these capabilities:
❶ Detect and register the number of people passing in and out of the camera frame.
❷ Generate a heat or motion map.
❸ Capture video and save snapshots.
How It Works
The application uses the captured video source and preprocesses the video frames through a heat map function, as well as a people-counting inference function. The results are then merged for the final output and stored both locally and in the cloud.
❶ Visualizing movement patterns over time starts with preprocessing the video frames using the HeatMap generation function and apply ColorMap to create a heat map.
❷ Using the inference engine in the Intel® Distribution of OpenVINO™ toolkit, it detects and counts the number of people within the video frame, and then draws a bounding box over each person detected.
❸ The video frames are merged to create the final output and are saved as snapshots locally. The snapshots can be uploaded to the cloud as a data blob using the Microsoft Azure* for Python* SDK.