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.
242 - 255
ECCV: European Conference on Computer Vision
2012-10-07 – 2012-10-13
Sharmanska V, Quadrianto N, Lampert C. Augmented attribute representations. In: Vol 7576. Springer; 2012:242-255. doi:10.1007/978-3-642-33715-4_18
Sharmanska, V., Quadrianto, N., & Lampert, C. (2012). Augmented attribute representations (Vol. 7576, pp. 242–255). Presented at the ECCV: European Conference on Computer Vision, Florence, Italy: Springer. https://doi.org/10.1007/978-3-642-33715-4_18
Sharmanska, Viktoriia, Novi Quadrianto, and Christoph Lampert. “Augmented Attribute Representations,” 7576:242–55. Springer, 2012. https://doi.org/10.1007/978-3-642-33715-4_18.
V. Sharmanska, N. Quadrianto, and C. Lampert, “Augmented attribute representations,” presented at the ECCV: European Conference on Computer Vision, Florence, Italy, 2012, vol. 7576, no. PART 5, pp. 242–255.
Sharmanska V, Quadrianto N, Lampert C. 2012. Augmented attribute representations. ECCV: European Conference on Computer Vision, LNCS, vol. 7576, 242–255.
Sharmanska, Viktoriia, et al. Augmented Attribute Representations. Vol. 7576, no. PART 5, Springer, 2012, pp. 242–55, doi:10.1007/978-3-642-33715-4_18.