Augmented attribute representations Conference Paper


Author(s): Sharmanska, Viktoriia; Quadrianto, Novi; Lampert, Christoph
Title: Augmented attribute representations
Title Series: LNCS
Affiliation IST Austria
Abstract: We propose a new learning method to infer a mid-level feature representation that combines the advantage of semantic attribute representations with the higher expressive power of non-semantic features. The idea lies in augmenting an existing attribute-based representation with additional dimensions for which an autoencoder model is coupled with a large-margin principle. This construction allows a smooth transition between the zero-shot regime with no training example, the unsupervised regime with training examples but without class labels, and the supervised regime with training examples and with class labels. The resulting optimization problem can be solved efficiently, because several of the necessity steps have closed-form solutions. Through extensive experiments we show that the augmented representation achieves better results in terms of object categorization accuracy than the semantic representation alone.
Keywords: Class labels; Optimization problems; Expressive power; Discriminative Autoencoder; Hybrid Representations; Closed form solutions; Learning methods; Mid-level features; Object categorization; Semantic attribute; Semantic representation; Smooth transitions; Training example
Conference Title: ECCV: European Conference on Computer Vision
Volume: 7576
Issue PART 5
Conference Dates: October 7 - 13, 2012
Conference Location: Florence, Italy
Publisher: Springer  
Date Published: 2012-10-01
Start Page: 242
End Page: 255
Sponsor: NQ is supported by a Newton International Fellowship.
DOI: 10.1007/978-3-642-33715-4_18
Open access: no
IST Austria Authors
  1. Christoph Lampert
    87 Lampert
Related IST Austria Work