How to compare faces in a Collection to faces in a Stored Video using AWS Rekognition?

问题

I would appreciate some guidance on the following issue.

Use Case:

• Create a collection with known faces.
• Search a stored video to identify "known" faces & draw a bounding box around them in the video frame

Steps taken:

• I'm able to create a collection and index faces I'm able to analyse the stored video and get the results of PersonMatch and FaceMatch using getFaceSearch()
• I'm able to draw the bounding boxes around Persons found in the video, etc., however...

Issue:

• The response of getFaceSearch() returns an array of FaceMatches.
• However, when I access the FaceMatch the coordinates are of the face found in the source image that was indexed in the collection, not of the face matched in the video.
• I've looked through the API documentation and have not been able to find any information on how to get the coordinates of a matched face in the video so I can draw a bounding box on the video frame. Here is the API document that I'm referring to. https://docs.aws.amazon.com/rekognition/latest/dg/API_Face.html

Thanks for your help on this issue!

来源：https://stackoverflow.com/questions/54961012/how-to-compare-faces-in-a-collection-to-faces-in-a-stored-video-using-aws-rekogn