Direction of human recognition based on video was made, in the feature extraction to extract feature codes of each frame? Used for classification is characterized by the signature of each frame? Have great spirit as a gesture recognition wants to know the main is don't understand programming implementation process

CodePudding user response:

On a related example code first

CodePudding user response:

For example you can find a link to send up to look at it, I think should be used only to extract key frames

CodePudding user response:

Think we should see the opencv