It is an annotation program that was used to annotate the Chinese Traffic Police Gesture Dataset v1.
In the Chinese Traffic Police Gesture Dataset v2, the LosslessCut is employed instead, due to more advanced cutting ability, flexible timeline and preview functions, memory usages etc. Instead of marking the start and end frame of a gesture, we use LosslessCut to mark the timestamp, then use opencv to find the corresponding frame.