https://pixabay.com/videos/cars-motorway-speed-motion-traffic-1900/ https://github.com/tensorflow/models/blob/master/res

[answerhappygod]

https://pixabay.com/videos/cars-motorwa ... ffic-1900/
https://github.com/tensorflow/models/bl ... ion_zoo.md
Use the SSD MobileNet V1 Coco (ssd_mobilenet_v1_coco) model. The video you'll process can be found on Pixabay. The 640x360 version of the video is smallest and easiest to handle, though any size should work since you must scale down the images for processing. Your program should: Read in a video file (use the one in this colab if you want) Load the TensorFlow model linked above • Loop over each frame of the video . • Scale the frame down to a size the model expects • Feed the frame to the model Loop over detections made by the model . If the detection score is above some threshold, draw a bounding box onto the frame and put a label in or near the box • Write the frame back to a new video .