Abstract: A headrest stay includes a detachment prevention slot portion formed in a rod-shaped leg portion, the detachment prevention slot portion being recessed toward a center side of the leg portion, the lock portion being configured to enter into the detachment prevention slot portion, and the detachment prevention slot portion including: a first flat face portion that faces a lower side; and a second flat face portion that faces an upper side, the second flat face portion including: an opposing portion that opposes the first flat face portion; and a region that, in a vertical sectional diagram cut along the axial direction of the leg portion, is disposed at an outer peripheral side of the leg portion relative to the first flat face portion.

Abstract: The present invention makes it possible to cause an action recognizer capable of recognizing actions with high accuracy and with a small quantity of learning data to learn. An input unit 101 receives input of a learning video and an action label indicating an action of an object, a detection unit 102 detects a plurality of objects included in each frame image included in the learning video, a direction calculation unit 103 calculates a direction of a reference object, which is an object to be used as a reference among the plurality of detected objects, a normalization unit 104 normalizes the learning video so that a positional relationship between the reference object and another object becomes a predetermined relationship, and an optimization unit 106 optimizes parameters of an action recognizer to estimate the action of the object in the inputted video based on the action estimated by inputting the normalized learning video to the action recognizer and the action indicated by the action label.

Abstract: An object is to accurately recognize an action of a subject. A direction alignment unit 24 is configured to perform at least one of rotation and inversion on an image based on an action direction of a desired subject in the image, so as to obtain an adjusted image. An action recognition device 26 is configured to recognize an action of the desired subject using the adjusted image as an input.