Robust Representation and Recognition of Facial Emotions Using Extreme Sparse Learning

Shojaeilangari, Seyedehsamaneh; Yau, Wei-Yun; Nandakumar, Karthik; Li, Jun; Teoh, Eam Khwang

doi:10.1109/tip.2015.2416634

Cited by 134 publications

(68 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several techniques have been proposed to solve the problem defined by Eq. , and can be divided into two broad classes: (a)The first group is based on optimization techniques that iterate between a representative set of sparse coefficient (estimated typically via Orthogonal Matching Pursuit, Basis Pursuit, Iterative Hard Thresholding, and etc gradient descent, and etc.).…”

Section: Methodsmentioning

confidence: 99%

“…(a)The first group is based on optimization techniques that iterate between a representative set of sparse coefficient (estimated typically via Orthogonal Matching Pursuit, 28 Basis Pursuit, 29 Iterative Hard Thresholding, 30 and etc. 31 ) and update of the dictionary using known sparse coefficients (utilizing algorithms such as K-Singular Value Decomposition, 32 gradient descent, 33 and etc.). In these techniques, the stopping criterion is defined by assuming knowledge of the noise variance or sparsity level of sparse coefficients a.…”

Section: A4 Dictionary Learning Based Inpaintingmentioning

confidence: 99%

See 1 more Smart Citation

Recovery of missing data in partial geometry PET scanners: Compensation in projection space vs image space

et al. 2018

Self Cite

View full text Add to dashboard Cite

Purpose: Robust and reliable reconstruction of images from noisy and incomplete projection data holds significant potential for proliferation of cost-effective medical imaging technologies. Since conventional reconstruction techniques can generate severe artifacts in the recovered images, a notable line of research constitutes development of appropriate algorithms to compensate for missing data and to reduce noise. In the present work, we investigate the effectiveness of state-of-the-art methodologies developed for image inpainting and noise reduction to preserve the quality of reconstructed images from under-sampled PET data. We aimed to assess and ascertain whether missing data recovery is best performed in the projection space prior to reconstruction or adjoined with the reconstruction step in image space. Methods: Different strategies for data recovery were investigated using realistic patient derived phantoms (brain and abdomen) in PET scanners with partial geometry (small and large gap structures). Specifically, gap filling strategies in projection space were compared with reconstruction based compensation in image space. The methods used for filling the gap structure in sinogram PET data include partial differential equation based techniques (PDE), total variation (TV) regularization, DCT-based penalized regression, and dictionary learning based inpainting (DLI). For compensation in image space, compressed sensing based image reconstruction methods were applied. These include the preconditioned alternating projection (PAPA) algorithm with first and higher order total variation (HOTV) regularization as well as dictionary learning based compressed sensing (DLCS). We additionally investigated the performance of the methods for recovery of missing data in presence of simulated lesion. The impact of different noise levels in the under-sampled sinograms on performance of the approaches were also evaluated. Results: In our first study (brain imaging), DLI was shown to outperform other methods for small gap structure in terms of root mean square error (RMSE) and structural similarity (SSIM), though having relatively high computational cost. For large gap structure, HOTV-PAPA produces better results. In the second study (abdomen imaging), again the best performance belonged to DLI for small gap, and HOTV-PAPA for large gap. In our experiments for lesion simulation on patient brain phantom data, the best performance in term of Contrast Recovery Coefficient (CRC) for small gap simulation belonged to DLI, while in the case of large gap simulation, HOTV-PAPA outperformed others. Our evaluation of the impact of noise on performance of approaches indicated that in case of low and medium noise levels, DLI still produces favorable results among inpainting approaches. However, for high noise levels, the performance of PDE4 (variant of PDE) and DLI are very competitive. Conclusions: Our results showed that estimation of missing data in projection space as a preprocessing step before reconstruction can improve the qual...

show abstract

Section: Methodsmentioning

confidence: 99%

Section: A4 Dictionary Learning Based Inpaintingmentioning

confidence: 99%

Recovery of missing data in partial geometry PET scanners: Compensation in projection space vs image space

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Both the above systems work only on static images. Seyedehsamaneh et.al [9] had proposed a system wherein the images are detected using optical flow method with Extreme Sparse Learning. This in our case of detecting only the negative emotions seems to be complex.…”

Section: Emotion Detectorsmentioning

confidence: 99%

Human Emotion Surveillance Using Computer Vision

Radha¹,

Atchatha.²,

Kaushik³

et al. 2018

IJET

View full text Add to dashboard Cite

India, a land of marvels, is outstanding in many aspects, its culture, ecosystem, etc. Sadly, it also ranks among the top countries in the world to have an annual suicide rate. This project aims at the foundation of human emotion surveillance. This system assists in the facial recognition, feature extraction and the threshold detection of stress for emotions expressed through face using the viola-jones algorithms and weak classifiers. This focuses basically on segregation of positive and negative emotions, detecting stress based on a usual threshold value and possibly providing an alternate means to let loose the extra stress built up if possible.

show abstract

“…The presence of emotions in the query image is detected by using an Extreme Sparse Learning (ESL) method [28] which uses a spatio-temporal descriptor to handle the pose invariant features. The spatio-temporal descriptor has been generated by means of concatenating the spatio-temporal features including facial expansion and contraction, local spins around the axis of facial muscles, projections, rotations etc.…”

Section: Robust Auxiliary Dictionary Learningmentioning

confidence: 99%

A Review of Face Recognition Techniques

Anesha¹

2016

JECSE

View full text Add to dashboard Cite

In today's networked world, the necessity of maintaining information safety as well as preserving the physical property is gradually becoming significant and tedious. As the requirement of high level of security arises, face recognition technology is bound to bend, to cope with the upcoming needs. Also, face recognition has caught the eye of researchers in the areas ranging from image processing and security to computer vision. Streams including content retrieval, network security and video compression get profited by face recognition technology. This is because face recognition deals with people as its centre of attention, thereby increasing the user-friendliness in human-computer interaction. It could also maintain our privacy and protect our assets without dropping our identity. But in this real world environment it seems to be a difficult task because of image variance in terms of position, size, expression and pose. In order to handle all these inter and intra-class variations in face images, various methodologies have been proposed. Here we analyse some methods, their possible strongholds and drawbacks.

show abstract

Robust Representation and Recognition of Facial Emotions Using Extreme Sparse Learning

Cited by 134 publications

References 38 publications

Recovery of missing data in partial geometry PET scanners: Compensation in projection space vs image space

Recovery of missing data in partial geometry PET scanners: Compensation in projection space vs image space

Human Emotion Surveillance Using Computer Vision

A Review of Face Recognition Techniques

Contact Info

Product

Resources

About