Attribute-based representation has shown great promises for visual recognition due to its intuitive interpretation and cross-category generalization property. However, human efforts are usually involved in the attribute designing process, making the representation costly to obtain. In this paper, we propose a novel formulation to automatically design discriminative "category-level attributes", which can be efficiently encoded by a compact category-attribute matrix. The formulation allows us to achieve intuitive and critical design criteria (category-separability, learnability) in a principled way. The designed attributes can be used for tasks of cross-category knowledge transfer, achieving superior performance over well-known attribute dataset Animals with Attributes (AwA) and a large-scale ILSVRC2010 dataset (1.2M images). This approach also leads to state-of-the-art performance on the zero-shot learning task on AwA.
Specifically, We propose a framework of using attributes as mid-level cues for multi-class classification. And the error of such classification scheme is used to measure the discriminativeness of attributes. The framework consists of two steps: attribute encoding, and category decoding.
Attribute-based query offers an intuitive way of image retrieval, in which users can describe the intended search targets with understandable attributes. In this paper, we develop a general and powerful framework to solve this problem by leveraging a large pool of weak attributes comprised of automatic classifier scores or other mid-level representations that can be easily acquired with little or no human labor. We extend the existing retrieval model of modeling dependency within query attributes to modeling dependency of query attributes on a large pool of weak attributes, which is more expressive and scalable. To efficiently learn such a large dependency model without overfitting, we further propose a semi-supervised graphical model to map each multi-attribute query to a subset of weak attributes. Through extensive experiments over several attribute benchmarks, we demonstrate consistent and significant performance improvements over the state-of-the-art techniques. In addition, we compile the largest multi-attribute image retrieval dateset to date, including 126 fully labeled query attributes and 6,000 weak attributes of 0.26 million images.
Felix X. Yu; Liangliang Cao; Rogerio S. Feris; John R. Smith; Shih-Fu Chang. Designing category-level attributes for discriminative visual recognition CVPR 2013 [PDF] [Supp] [Github]
Felix X. Yu; Liangliang Cao; Rogerio S. Feris; John R. Smith; Shih-Fu Chang. Additional remarks on designing category-level attributes for discriminative visual recognition Technical Report # CUCS 007-13[PDF]
Felix X. Yu; Rongrong Ji; Ming-Hen Tsai; Guangnan Ye; Shih-Fu Chang. Weak attributes for large-scale image retrieval CVPR 2012 [PDF]
Felix X. Yu; Rongrong Ji; Ming-Hen Tsai; Guangnan Ye; Shih-Fu Chang. Experiments of image retrieval using weak attributes Technical Report # CUCS 005-12 [PDF]