Label-Embedding for Attribute-Based Classification - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2013

Label-Embedding for Attribute-Based Classification


Attributes are an intermediate representation, which enables parameter sharing between classes, a must when training data is scarce. We propose to view attribute-based image classification as a label-embedding problem: each class is embedded in the space of attribute vectors. We introduce a function which measures the compatibility between an image and a label embedding. The parameters of this function are learned on a training set of labeled samples to ensure that, given an image, the correct classes rank higher than the incorrect ones. Results on the Animals With Attributes and Caltech-UCSD-Birds datasets show that the proposed framework outperforms the standard Direct Attribute Prediction baseline in a zero-shot learning scenario. The label embedding framework offers other advantages such as the ability to leverage alternative sources of information in addition to attributes (e.g. class hierarchies) or to transition smoothly from zero-shot learning to learning with large quantities of data.
Fichier principal
Vignette du fichier
PID2749989.pdf (2.79 Mo) Télécharger le fichier
Vignette du fichier
methodology.png (193.46 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Format : Figure, Image

Dates and versions

hal-00815747 , version 1 (19-04-2013)
hal-00815747 , version 2 (22-04-2013)
hal-00815747 , version 3 (11-07-2013)



Zeynep Akata, Florent Perronnin, Zaid Harchaoui, Cordelia Schmid. Label-Embedding for Attribute-Based Classification. CVPR - IEEE Computer Vision and Pattern Recognition, Jun 2013, Portland, United States. pp.819-826, ⟨10.1109/CVPR.2013.111⟩. ⟨hal-00815747v3⟩
1355 View
6042 Download



Gmail Facebook X LinkedIn More