Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/107686
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | Robust optimization for deep regression |
Author: | Belagiannis, V. Rupprecht, C. Carneiro, G. Navab, N. |
Citation: | Proceedings / IEEE International Conference on Computer Vision. IEEE International Conference on Computer Vision, 2015, vol.2015 International Conference on Computer Vision, ICCV 2015, pp.2830-2838 |
Publisher: | IEEE |
Issue Date: | 2015 |
Series/Report no.: | IEEE International Conference on Computer Vision |
ISBN: | 9781467383912 |
ISSN: | 1550-5499 |
Conference Name: | 2015 IEEE International Conference on Computer Vision (ICCV 2015) (11 Dec 2015 - 18 Dec 2015 : Santiago, CHILE) |
Statement of Responsibility: | Vasileios Belagiannis, Christian Rupprecht, Gustavo Carneiro, and Nassir Navab |
Abstract: | Convolutional Neural Networks (ConvNets) have successfully contributed to improve the accuracy of regressionbased methods for computer vision tasks such as human pose estimation, landmark localization, and object detection. The network optimization has been usually performed with L2 loss and without considering the impact of outliers on the training process, where an outlier in this context is defined by a sample estimation that lies at an abnormal distance from the other training sample estimations in the objective space. In this work, we propose a regression model with ConvNets that achieves robustness to such outliers by minimizing Tukey's biweight function, an M-estimator robust to outliers, as the loss function for the ConvNet. In addition to the robust loss, we introduce a coarse-to-fine model, which processes input images of progressively higher resolutions for improving the accuracy of the regressed values. In our experiments, we demonstrate faster convergence and better generalization of our robust loss function for the tasks of human pose estimation and age estimation from face images. We also show that the combination of the robust loss function with the coarse-to-fine model produces comparable or better results than current state-of-the-art approaches in four publicly available human pose estimation datasets. |
Keywords: | Training, robustness, machine learning, convergence, minimization |
Rights: | © 2015 IEEE |
DOI: | 10.1109/ICCV.2015.324 |
Grant ID: | http://purl.org/au-research/grants/arc/CE140100016 |
Published version: | http://dx.doi.org/10.1109/iccv.2015.324 |
Appears in Collections: | Aurora harvest 8 Computer Science publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
RA_hdl_107686.pdf Restricted Access | Restricted Access | 745.44 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.