Learning the Roots of Visual Domain Shift

In this paper we focus on the spatial nature of visual domain shift, attempting to learn where domain adaptation originates in each given image of the source and target set. We borrow concepts and techniques from the CNN visualization literature, and learn domainness maps able to localize the degree of domain specificity in images. We derive from these maps features related to different domainness levels, and we show that by considering them as a preprocessing step for a domain adaptation algorithm, the final classification performance is strongly improved. Combined with the whole image representation, these features provide state of the art results on the Office dataset.

[1]  Ilja Kuzborskij,et al.  When Naïve Bayes Nearest Neighbors Meet Convolutional Neural Networks , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Trevor Darrell,et al.  Deep Domain Confusion: Maximizing for Domain Invariance , 2014, CVPR 2014.

[3]  Bernhard Schölkopf,et al.  A Kernel Two-Sample Test , 2012, J. Mach. Learn. Res..

[4]  Trevor Darrell,et al.  Discovering Latent Domains for Multisource Domain Adaptation , 2012, ECCV.

[5]  Trevor Darrell,et al.  Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Tinne Tuytelaars,et al.  Unsupervised Visual Domain Adaptation Using Subspace Alignment , 2013, 2013 IEEE International Conference on Computer Vision.

[7]  Yuan Shi,et al.  Geodesic flow kernel for unsupervised domain adaptation , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[8]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[9]  Trevor Darrell,et al.  Efficient Learning of Domain-invariant Image Representations , 2013, ICLR.

[10]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[11]  Michael I. Jordan,et al.  Learning Transferable Features with Deep Adaptation Networks , 2015, ICML.

[12]  Bolei Zhou,et al.  Object Detectors Emerge in Deep Scene CNNs , 2014, ICLR.

[13]  Trevor Darrell,et al.  Simultaneous Deep Transfer Across Domains and Tasks , 2015, ICCV.

[14]  Ilja Kuzborskij,et al.  When Na\"ive Bayes Nearest Neighbours Meet Convolutional Neural Networks , 2015, 1511.03853.

[15]  Antonio Torralba,et al.  Transfer Learning by Borrowing Examples for Multiclass Object Detection , 2011, NIPS.

[16]  Joshua B. Tenenbaum,et al.  Learning to share visual appearance for multiclass object detection , 2011, CVPR 2011.

[17]  Ivor W. Tsang,et al.  Domain Transfer SVM for video concept detection , 2009, CVPR 2009.

[18]  Barbara Caputo,et al.  Frustratingly Easy NBNN Domain Adaptation , 2013, 2013 IEEE International Conference on Computer Vision.

[19]  Trevor Darrell,et al.  Adapting Visual Category Models to New Domains , 2010, ECCV.

[20]  Kristen Grauman,et al.  Connecting the Dots with Landmarks: Discriminatively Learning Domain-Invariant Features for Unsupervised Domain Adaptation , 2013, ICML.

[21]  Kristen Grauman,et al.  Reshaping Visual Datasets for Domain Adaptation , 2013, NIPS.

[22]  Svetlana Lazebnik,et al.  Multi-scale Orderless Pooling of Deep Convolutional Activation Features , 2014, ECCV.

[23]  Ivor W. Tsang,et al.  Domain adaptation from multiple sources via auxiliary classifiers , 2009, ICML '09.

[24]  Tinne Tuytelaars,et al.  Joint cross-domain classification and subspace learning for unsupervised adaptation , 2014, Pattern Recognit. Lett..

[25]  Andrew Zisserman,et al.  Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps , 2013, ICLR.