Abstract
Although videos appear to be very high-dimensional in terms of duration × frame-rate × resolution, temporal smoothness constraints ensure that the intrinsic dimensionality for videos is much lower. In this paper, we use this idea for investigating Domain Adaptation (DA) in videos, an area that remains under-explored. An approach that has worked well for the image DA is based on the subspace modeling of the source and target domains, which works under the assumption that the two domains share a latent subspace where the domain shift can be reduced or eliminated. In this paper, first we extend three subspace based image DA techniques for human action recognition and then combine it with our proposed Eclectic Domain Mixing (EDM) approach to improve the effectiveness of the DA. Further, we use discrepancy measures such as Symmetrized KL Divergence and Target Density Around Source for empirical study of the proposed EDM approach. While, this work mainly focuses on Domain Adaptation in videos, for completeness of the study, we comprehensively evaluate our approach using both object and action datasets. In this paper, we have achieved consistent improvements over chosen baselines and obtained some state-of-the-art results for the datasets.
Similar content being viewed by others
References
Aljundi R, Emonet R, Muselet D, Sebban M (2015) Landmarks-based kernelized subspace alignment for unsupervised domain adaptation. In: CVPR
Baktashmotlagh M, Harandi M, Lovell B, Salzmann M (2013) Unsupervised domain adaptation by domain invariant projection. In: ICCV
Baktashmotlagh M, Harandi MT, Lovell BC, Salzmann M (2014) Domain adaptation on the statistical manifold. In: CVPR
Ben-David S, Blitzer J, Crammer K, Pereira F (2007) Analysis of representations for domain adaptation. In: NIPS, pp 137–144
Bergamo A, Torresani L (2010) Exploiting weakly-labeled web images to improve object classification: a domain adaptation approach. In: NIPS
Blitzer J, McDonald R, Pereira F (2006) Domain adaptation with structural correspondence learning. In: Proc of EMNLP, pp 120–128
Caseiro R, Henriques JF, Martins P, Batista J (2015) Beyond the shortest path Unsupervised domain adaptation by Sampling Subspaces along the Spline Flow. In: CVPR
Cross-Dataset Setup: https://sites.google.com/site/crossdataset/
Csurka G (2017) Domain adaptation for visual applications: a comprehensive survey. In: Arxiv
Daume IIIH (2007) Frustratingly easy domain adaptation. In: Proc of ACL, pp 256–263
Daume IIIH, Kumar A, Saha A (2010) Co-regularization based semi-supervised domain adaptation. In: NIPS
Daume IIIH, Marcu D (2006) Domain adaptation for statistical classifiers. JAIR
Donahue J, Hendricks LA, Guadarrama S, Rohrbach M, Venugopalan S, Saenko K, Darrell T (2015) Long-term recurrent convolutional networks for visual recognition and description. In: CVPR
Duan L, Tsang I, Xu D, Chua T (2009) Domain adaptation from multiple sources via auxiliary classifiers. In: ICML
Duan L, Tsang I, Xu D, Maybank S (2009) Domain transfer SVM for video concept detection. In: CVPR
Duan L, Xu D, Tsang IW (2012) Domain adaptation from multiple Sources: a Domain-Dependent regularization approach. In: IEEE Transactions on neural networks and learning systems
Duan L, Xu D, Tsang IW, Luo J (2012) Visual event recognition in videos by learning from web data. In: IEEE Transactions on pattern analysis and machine intelligence
Faraji Davar N, deCampos TE, Windridge D, Kittler J, Christmas W (2011) Domain adaptation in the context of sport video action recognition. In: Domain adaptation workshop in conjunction with NIPS
Feichtenhofer C, Pinz A, Zisserman A (2016) Convolutional two-stream network fusion for video action recognition. In: Arxiv
Fernando B, Gavves EM, Jos O, Ghodrati A, Tuytelaars T (2015) Modeling video evolution for action recognition. In: CVPR
Fernando B, Habrard A, Sebban M, Tuytelaars T (2013) Unsupervised visual domain adaptation using subspace alignment. In: ICCV
Ganin Y, Lempitsky VS (2015) Unsupervised domain adaptation by backpropagation. In: ICML
Gong B, Grauman K, Sha F (2013) Connecting the dots with landmarks discriminatively learning domain-invariant features for unsupervised domain adaptation. In: ICML
Gong B, Shi Y, Sha F, Grauman K (2012) Geodesic flow kernel for unsupervised domain adaptation. In: CVPR
Gopalan R, Li R, Chellappa R (2014) Unsupervised adaptation across domain shifts by generating intermediate data representations. In: IEEE Trans on pattern anal Mach Intell
Gopalan R, Li R, Patel V, Chellappa R (2015) Domain adaptation for visual recognition. In: Found Trends Comput Graph Vis
Hoffman J, Kulis B, Darrell T, Saenko K (2012) Discovering latent domains for multi-source domain adaptation. In: ECCV
Jiang F, Zhang S, Wu S, Gao Y, Zhao D (2015) Multi-layered gesture recognition with Kinect. J Mach Learn Res 16:227–254
KTH and MSR Action II Dataset: http://www.cs.utexas.edu/~chaoyeh/web_action_data/dataset_list.html
Kuhne H, Jhuang H, Garrote E, Poggio T, Serre T (2011) HMDB a large video database for human motion recognition. In: ICCV
Kulis B, Saenko K, Darrell T (2011) What you saw is not what you get Domain adaptation using asymmetric kernel transforms. In: Proc of CVPR, pp 1785–1792
Li Ruonan (2012) Discriminative virtual views for cross-view action recognition. In: CVPR
Long M, Cao Y, Wang J, Jordan MI (2015) Learning transferable features with deep adaptation networks. In: ICML
Long M, Wang J, Jordan MI (2016) Unsupervised domain adaptation with residual transfer networks. In: Arxiv
Long M, Zhu H, Wang J, Jordan MI (2017) Deep transfer learning with joint adaptation networks. In: Arxiv
Ni J, Qiu Q, Chellappa R (2013) Subspace interpolation via dictionary learning for unsupervised domain adaptation. In: CVPR
Niebles JC, Chen C-W, Fei-Fei L (2010) Modeling temporal structure of decomposable motion segments for activity classification. In: ECCV
Pan S, Tsang I, Kwok J, Yang Q (2009) Domain adaptation via transfer component analysis. IEEE Trans Neural Nets 99:1–12
Qiu Q, Patel V, Turaga P, Chellappa R (2012) Domain adaptive dictionary learning. In: ECCV
Reddy KK, Shah M (2013) Recognizing 50 human action categories of web videos. In: Mach Vision appl
Saenko K, Kulis B, Fritz M, Darrell T (2010) Adapting visual category models to new domains. In: ECCV
Simonyan K, Zisserman A (2014) Two-Stream Convolutional networks for action rec. In: NIPS
Smola A, Gretton A, Song L, Schölkopf B (2007) A Hilbert space embedding for distributions. In: Algorithmic learning theory
Sultani W, Saleemi I (2014) Human action recognition across datasets by Foreground-Weighted histogram decomposition. In: CVPR
Sun B, Feng J, Saenko K (2016) Return of frustratingly easy domain adaptation. In: Conference on artificial intelligence
Sun B, Saenko K (2015) Subspace distribution alignment for unsupervised domain adaptation. In: BMVC
Tommasi T, Tuytelaars T (2014) A testbed for cross-dataset analysis. In: ECCV
Tran D u, Bourdev LD, Fergus R, Torresani L, Paluri M (2015) Learning Spatio-temporal features with 3D convolutional networks. In: ICCV
Tzeng E, Hoffman T, Darrell J, Saenko K (2015) Simultaneous deep transfer across domains and tasks. In: Arxiv
Wang H, Schmid C (2013) Action recognition with improved trajectories. In: Proc. ICCV, pp 3551–3558
Zhang S, Yao H, Sun X, Wang K, Zhang J, Lu X, Zhang Y (2014) Action recognition based on overcomplete independent components analysis. Inf Sci 281:635–647
Zhang Z, Wang C, Xiao B, Zhou W, Liu S, Shi C (2013) Cross-View Action recognition via a continuous virtual path. In: CVPR
Acknowledgments
The authors would like to thank Director, Centre for AI & Robotics, Bangalore for permitting the publication of research work.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Jamal, A., Deodhare, D., Namboodiri, V. et al. Eclectic domain mixing for effective adaptation in action spaces. Multimed Tools Appl 77, 29949–29969 (2018). https://doi.org/10.1007/s11042-018-6179-y
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-018-6179-y