Robust coupling in space of sparse codes for multi-view recognition

Taalimi, Ali; Rahimpour, Alireza; Capdevila, Cristian; Zhang, Zhifei; Qi, Hairong

doi:10.1109/icip.2016.7533090

Cited by 13 publications

(9 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The corresponding dictionary of each view has one normalized image from each subject in that view, p = 20. MT-VSN learns a dictionary with uncorrelated class-specific atoms and outperforms [30,32] by more than 5% and enhances [26] more than 2%.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Multi-view task-driven recognition in visual sensor networks

Taalimi

Rahimpour

Liu

et al. 2017

2017 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

Nowadays, distributed smart cameras are deployed for a wide set of tasks in several application scenarios, ranging from object recognition, image retrieval, and forensic applications. Due to limited bandwidth in distributed systems, efficient coding of local visual features has in fact been an active topic of research. In this paper, we propose a novel approach to obtain a compact representation of high-dimensional visual data using sensor fusion techniques. We convert the problem of visual analysis in resource-limited scenarios to a multiview representation learning, and we show that the key to finding properly compressed representation is to exploit the position of cameras with respect to each other as a normbased regularization in the particular signal representation of sparse coding. Learning the representation of each camera is viewed as an individual task and a multi-task learning with joint sparsity for all nodes is employed. The proposed representation learning scheme is referred to as the multi-view task-driven learning for visual sensor network (MT-VSN). We demonstrate that MT-VSN outperforms state-of-the-art in various surveillance recognition tasks.

show abstract

Section: Methodsmentioning

confidence: 99%

“…The A i 12 promotes a solution with sparse non-zero rows. Hence, similar support is enforced on A i at the column level of each dictionary D m [26].…”

Section: Multi-view Task Driven Dictionary Learningmentioning

confidence: 99%

Multi-view task-driven recognition in visual sensor networks

Taalimi

Rahimpour

Liu

et al. 2017

2017 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…A dictionary in signal processing can be simplify as a set of fixed variables and then seeks for the solution as a linear combination of variables in the dictionary. The dictionary should be designed so that it can successfully generalize unseen and new data [9]. The prior information about the data or the form of the solution leads to the concept of regularization, which shows promising performance in dealing with unseen data.…”

Section: Introductionmentioning

confidence: 99%

Event analysis of pulse-reclosers in distribution systems through sparse representation

Raoufat

Taalimi

Tomsovic

et al. 2017

2017 19th International Conference on Intelligent System Application to Power Systems (ISAP)

Self Cite

View full text Add to dashboard Cite

The pulse-recloser uses pulse testing technology to verify that the line is clear of faults before initiating a reclose operation, which significantly reduces stress on the system components (e.g. substation transformers) and voltage sags on adjacent feeders. Online event analysis of pulse-reclosers are essential to increases the overall utility of the devices, especially when there are numerous devices installed throughout the distribution system. In this paper, field data recorded from several devices were analyzed to identify specific activity and fault locations. An algorithm is developed to screen the data to identify the status of each pole and to tag time windows with a possible pulse event. In the next step, selected time windows are further analyzed and classified using a sparse representation technique by solving an 1-regularized least-square problem. This classification is obtained by comparing the pulse signature with the reference dictionary to find a set that most closely matches the pulse features. This work also sheds additional light on the possibility of fault classification based on the pulse signature. Field data collected from a distribution system are used to verify the effectiveness and reliability of the proposed method.

show abstract

“…In this work we propose a sparse decomposition based segmentation, which tries to resolve these problems. Sparse representation has been used for various applications in recent years, including face recognition , super-resolution, morphological component analysis, image restoration, image denoising and sparse coding [17]- [26]. Within our sparse decomposition framework, we also impose suitable priors on each layer, in particular smoothness on the background, and connectivity on the foreground.…”

Section: Introductionmentioning

confidence: 99%

Image segmentation using overlapping group sparsity

Minaee

Wang

2016

2016 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)

View full text Add to dashboard Cite

Abstract-Sparse decomposition has been widely used for different applications, such as source separation, image classification and image denoising. This paper presents a new algorithm for segmentation of an image into background and foreground text and graphics using sparse decomposition. First, the background is represented using a suitable smooth model, which is a linear combination of a few smoothly varying basis functions, and the foreground text and graphics are modeled as a sparse component overlaid on the smooth background. Then the background and foreground are separated using a sparse decomposition framework and imposing some prior information, which promote the smoothness of background, and the sparsity and connectivity of foreground pixels. This algorithm has been tested on a dataset of images extracted from HEVC standard test sequences for screen content coding, and is shown to outperform prior methods, including least absolute deviation fitting, k-means clustering based segmentation in DjVu, and shape primitive extraction and coding algorithm.

show abstract

Robust coupling in space of sparse codes for multi-view recognition

Cited by 13 publications

References 19 publications

Multi-view task-driven recognition in visual sensor networks

Multi-view task-driven recognition in visual sensor networks

Event analysis of pulse-reclosers in distribution systems through sparse representation

Image segmentation using overlapping group sparsity

Contact Info

Product

Resources

About