APPARATUS FOR DETERMINING OBJECT TO BE VISUALLY RECOGNIZED

To improve accuracy of determining an object to be visually recognized at which a driver of a vehicle looks.SOLUTION: An apparatus for determining an object to be visually recognized includes: a line-of -sight detection unit which detects a line-of-sight direction of a drive of a vehicle; a vehicle information acquisition unit which acquires vehicle information indicating a travel state of the vehicle; an image acquisition unit which acquires a captured image obtained by an imaging unit imaging the surroundings of the vehicle; a gaze area object extraction unit which extracts a gaze area in the captured image at which the driver looks, on the basis of a result of detecting the line-of-sight direction; a candidate detection unit which recognizes objects included in the captured image, generates a top-down saliency map of the captured image on the basis of the captured image and the vehicle information, and detects objects having saliency in the top-down saliency map, out of the recognized objects, as candidates of an object to be visually recognized at which the driver looks; and an object determination unit which determines an object to be visually recognized at which the driver looks, on the basis of a result of extracting a gaze area and a result of detecting candidates of the object to be visually recognized.SELECTED DRAWING: Figure 4 【課題】車両の運転者の視線が向いている視認対象物の判定精度を向上させることが可能とする。【解決手段】実施形態の視認対象判定装置は、一例として、車両の運転者の視線方向を検出する視線検出部と、車両の走行状態を示す車両情報を取得する車両情報取得部と、撮像部によって車両の周辺を撮像して得られる撮像画像を取得する画像取得部と、視線方向の検出結果に基づいて、撮像画像内において運転者の視線が向けられている注視領域を抽出する注視領域物抽出部と、撮像画像に含まれる物体を認識し、撮像画像および車両情報に基づいて、撮像画像のトップダウンサリエンシーマップを生成し、認識された物体のうち、トップダウンサリエンシーマップにおいてサリエンシーを有する物体を、運転者の視線が向けられている視認対象物の候補として検出する候補検出部と、注視領域の抽出結果および視認対象物の候補の検出結果に基づいて、運転者の視線が向けられている視認対象物を判定する視認対象判定部と、を備える。【選択図】図４