Methods and systems to automatically record relevant action in a gaming environment

Systems and methods are described for detecting a triggering event in a gaming environment and recording a segment of video. In some aspects, the method may include obtaining a camera projection matrix associated with the gaming environment that correlates points in the gaming environment between a two-dimensional pixel space and a three-dimensional physical representation. An input video may be obtained from the gaming environment. Pose coordinates, including two-dimensional pixel space location, of a body part of a person in the input video may be estimated, and converted to three-dimensional pose coordinates using the camera projection matrix. A person of interest having a body part located within a region of interest in the gaming environment may be identified to determine if a triggering event has occurred. Storage of an input video segment that includes the trigger movement may be initiated on a storage device in communication with the user device.