Recognition for objects by relationship between attributes
Hiroka Horiguchi, Kazuo Ikeshiro, Hiroki Imamura · 2016
The object recognition method based on attributes has been studied. The conventional method recognizes objects by the presence or absence of attributes. However, the conventional method has two problems. Firstly, the conventional method is not able to recognize a target object of which a part of attribute is occluded. Secondly, the conventional method misrecognizes a target object which has irrelevant attributes. Therefore, to solve these two problems, we propose the object recognition by relationship between attributes. In this paper, we focus on the face as recognition object. The proposed method uses relationship as constraints for object recognition using attributes. The proposed method applies two major type constraints. First constraint is a local constraint which is applied to a part of attributes. To achieve robust face recognition against occlusion scenes, the proposed method uses the local constraint. And then, Second constraint is a global constraint which is applied to all attributes. To achieve robust face recognition against irrelevant attributes, the proposed method uses the global constraint. In this paper, to evaluate effectiveness of the proposed method, we compare the proposed method with the conventional method. We experimented in normal face, occlusion and irrelevant attributes. We used 43 images of a face which are changed in scale and rotation. Experimental results showed that the recognition ratio of the proposed method is equal to or more than the conventional method in normal face, occlusion and irrelevant attributes.