Publication
Optimization of robust loss functions for weakly-labeled image taxonomies: An ImageNet case study
Conference Article
Conference
International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR)
Edition
8th
Pages
355-368
Doc link
http://dx.doi.org/10.1007/978-3-642-23094-3_26
File
Abstract
The recently proposed ImageNet dataset consists of several million images, each annotated with a single object category. However, these annotations may be imperfect, in the sense that many images contain multiple objects belonging to the label vocabulary. In other words, we have a multi-label problem but the annotations include only a single label (and not necessarily the most prominent). Such a setting motivates the use of a robust evaluation measure, which allows for a limited number of labels to be predicted and, as long as one of the predicted labels is correct, the overall prediction should be considered correct. This is indeed the type of evaluation measure used to assess algorithm performance in a recent competition on ImageNet data. Optimizing such types of performance measures presents several hurdles even with existing structured output learning methods. Indeed, many of the current state-of-the-art methods optimize the prediction of only a single output label, ignoring this ‘structure’ altogether. In this paper, we show how to directly optimize continuous surrogates of such performance measures using structured output learning techniques with latent variables. We use the output of existing binary classifiers as input features in a new learning stage which optimizes the structured loss corresponding to the robust performance measure. We present empirical evidence that this allows us to ‘boost’ the performance of existing binary classifiers which are the state-of-the-art for the task of object classification in ImageNet.
Categories
image classification.
Author keywords
imagenet, image annotation
Scientific reference
J.J. McAuley, A. Ramisa and T.S. Caetano. Optimization of robust loss functions for weakly-labeled image taxonomies: An ImageNet case study, 8th International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition, 2011, St. Petersburg, Russia, in Energy Minimization Methods in Computer Vision and Pattern Recognition, Vol 6819 of Lecture Notes in Computer Science, pp. 355-368, 2011, Springer.
Follow us!