Enhancing Action Recognition by Cross-Domain Dictionary Learning

Zhu, Fan; Shao, Ling

doi:10.5244/c.27.52

Cited by 21 publications

(14 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The function C(·) is defined as the distance in the new sparse space of original nearest couples, a small C(·) indicates the data maintain more relationship in new sparse space. This idea is inspired by [22,23], in their method, this function is designed to measure the distances of similar crossdomain instances of the same class. However, our method is exactly unsupervised and directly perform on low-level feature.…”

Section: Unsupervised Domain Adaption Dictionary Learningmentioning

confidence: 99%

“…Huang and Wang [21] proposed a joint model which learns a pair of dictionaries with a feature space for describing and associating cross-domain data. In [22,23], Zhu and Shao proposed a weakly-supervised framework learns a pairwise dictionaries and a classifier while considering the capacity of the dictionaries in terms of reconstructability, discriminability and domain adaptability.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Domain Adaptation Dictionary Learning for Visual Recognition

Zhong

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Over the last years, dictionary learning method has been extensively applied to deal with various computer vision recognition applications, and produced state-of-the-art results. However, when the data instances of a target domain have a different distribution than that of a source domain, the dictionary learning method may fail to perform well. In this paper, we address the cross-domain visual recognition problem and propose a simple but effective unsupervised domain adaption approach, where labeled data are only from source domain. In order to bring the original data in source and target domain into the same distribution, the proposed method forcing nearest coupled data between source and target domain to have identical sparse representations while jointly learning dictionaries for each domain, where the learned dictionaries can reconstruct original data in source and target domain respectively. So that sparse representations of original data can be used to perform visual recognition tasks. We demonstrate the effectiveness of our approach on standard datasets. Our method performs on par or better than competitive state-ofthe-art methods.

show abstract

Section: Unsupervised Domain Adaption Dictionary Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Unsupervised Domain Adaptation Dictionary Learning for Visual Recognition

Zhong

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…We review these works from both dictionary learning and transfer learning (a.k.a., domain adaption, domain transfer or knowledge transfer) aspects. Learning an over-complete dictionary for sparse coding has been applied to various areas in computer vision and artificial intelligence, for instance, image restoration [9], image denoising [11], and action recognition [10]. The K-Singular Value Decomposition (K-SVD) [6] method, as a classical solution to 0 -based dictionary learning, focuses on the reconstruction capability of the learned dictionary.…”

Section: Related Workmentioning

confidence: 99%

Color object recognition via cross-domain learning on RGB-D images

Huang

Zhu

Shao

et al. 2016

2016 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

This paper addresses the object recognition problem using multiple-domain inputs. We present a novel approach that utilizes labeled RGB-D data in the training stage, where depth features are extracted for enhancing the discriminative capability of the original learning system that only relies on RGB images. The highly dissimilar source and target domain data are mapped into a unified feature space through transfer at both feature and classifier levels. In order to alleviate cross-domain discrepancy, we employ a state-of-the-art domain-adaptive dictionary learning algorithm that updates image representations in both domains and the classifier parameters simultaneously. The proposed method is trained on a RGB-D Object dataset and evaluated on the Caltech-256 dataset. Experimental results suggest that our approach can lead to significant performance gain over the state-of-the-art methods

show abstract

“…Although this is sometimes denoted domain adaptation, the latter usually refers to methods that regularize the target feature space, rather than the models themselves. This is frequently implemented by learning a feature transformation that maximizes the similarity of feature vectors from target and auxiliary domains [71][72][73]41,74,75]. Some methods have also been proposed to implement both domain and model adaptation [42].…”

Section: Related Workmentioning

confidence: 99%

Cross-modal domain adaptation for text-based regularization of image semantics in image retrieval systems

Pereira

Vasconcelos

2014

Computer Vision and Image Understanding

View full text Add to dashboard Cite

a b s t r a c tIn query-by-semantic-example image retrieval, images are ranked by similarity of semantic descriptors. These descriptors are obtained by classifying each image with respect to a pre-defined vocabulary of semantic concepts. In this work, we consider the problem of improving the accuracy of semantic descriptors through cross-modal regularization, based on auxiliary text. A cross-modal regularizer, composed of three steps, is proposed. Training images and text are first mapped to a common semantic space. A regularization operator is then learned for each concept in the semantic vocabulary. This is an operator which maps the semantic descriptors of images labeled with that concept to the descriptors of the associated texts. A convex formulation of the learning problem is introduced, enabling the efficient computation of concept-specific regularization operators. The third step is the selection of the most suitable operator for the image to regularize. This is implemented through a quantization of the semantic space, where a regularization operator is associated with each quantization cell. Overall, the proposed regularizer is a non-linear mapping, implemented as a piecewise linear transformation of the semantic image descriptors to regularize. This transformation is a form of cross-modal domain adaptation. It is shown to achieve better performance than recent proposals in the domain adaptation literature, while requiring much simpler optimization.

show abstract

Enhancing Action Recognition by Cross-Domain Dictionary Learning

Cited by 21 publications

References 31 publications

Unsupervised Domain Adaptation Dictionary Learning for Visual Recognition

Unsupervised Domain Adaptation Dictionary Learning for Visual Recognition

Color object recognition via cross-domain learning on RGB-D images

Cross-modal domain adaptation for text-based regularization of image semantics in image retrieval systems

Contact Info

Product

Resources

About