A Bayesian approach to adaptive video super resolution

Liu, Ce; Sun, Deqing

doi:10.1109/cvpr.2011.5995614

Cited by 236 publications

(213 citation statements)

References 27 publications

Supporting

Mentioning

213

Contrasting

Order By: Relevance

“…This architecture provides better performance and increases high amount of accuracy. This SCA (Sparse Coding Based Architecture) implemented in corporation with neural networks to reconstruct a high resolution image from the original low-resolution image using LIST (Learned Iterative Shrinkage and Thresholding) approach [40]. The Fig.2 shows the architectural diagram of Sparse Coding Based Architecture (SCA).…”

Section: Sparse Coding Reconstructionmentioning

confidence: 99%

“…Bayesian Adaptive Video Super Resolution model presented to get better high resolution reconstructed image with great feature extractions and performance degradation issues occurs whenever scaling factor increases. In [40], learning a Mixture of Deep Networks for Single Image Super-Resolution model proposed which contains ill-posed, complex mapping of low-resolution images and inverse image recovery problems.…”

Section: Video Scaling Issuesmentioning

confidence: 99%

See 1 more Smart Citation

Real-Time Video Scaling Based on Convolution Neural Network Architecture

Safinaz¹,

Kumar²

2017

IJIVP

View full text Add to dashboard Cite

In recent years, video super resolution techniques becomes mandatory requirements to get high resolution videos. Many super resolution techniques researched but still video super resolution or scaling is a vital challenge. In this paper, we have presented a real-time video scaling based on convolution neural network architecture to eliminate the blurriness in the images and video frames and to provide better reconstruction quality while scaling of large datasets from lower resolution frames to high resolution frames. We compare our outcomes with multiple exiting algorithms. Our

show abstract

Section: Sparse Coding Reconstructionmentioning

confidence: 99%

Section: Video Scaling Issuesmentioning

confidence: 99%

Real-Time Video Scaling Based on Convolution Neural Network Architecture

Safinaz¹,

Kumar²

2017

IJIVP

View full text Add to dashboard Cite

show abstract

“…Methods based on frequency domain are usually using the Fouier transform and wavelet transform (S. A. Devi et al, 2012;Rasti P et al, 2014) Methods based on spatial mainly includes: interpolation method (Batz M et al, 2015;Makwana R R, et al, 2013), regularization method (Yinhui L I et al, 2015;JM Fadili et al, 2009) and the MAP method (Villena S et al, 2013;Liu C et al, 2011) and so on. The super resolution reconstruction methods have achieve some progress and breakthrough, but at the same time the algorithm with highly complexity, the model of noise simple, poor real-time performance and robustness, and the detail information of result is not remarkable.…”

Section: Introductionmentioning

confidence: 99%

Super Resolution Reconstruction Based on Adaptive Detail Enhancement for Zy-3 Satellite Images

Zhu¹,

Song²,

Tan³

et al. 2016

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Super-resolution reconstruction of sequence remote sensing image is a technology which handles multiple low-resolution satellite remote sensing images with complementary information and obtains one or more high resolution images. The cores of the technology are high precision matching between images and high detail information extraction and fusion. In this paper puts forward a new image super resolution model frame which can adaptive multi-scale enhance the details of reconstructed image. First, the sequence images were decomposed into a detail layer containing the detail information and a smooth layer containing the large scale edge information by bilateral filter. Then, a texture detail enhancement function was constructed to promote the magnitude of the medium and small details. Next, the non-redundant information of the super reconstruction was obtained by differential processing of the detail layer, and the initial super resolution construction result was achieved by interpolating fusion of non-redundant information and the smooth layer. At last, the final reconstruction image was acquired by executing a local optimization model on the initial constructed image. Experiments on ZY-3 satellite images of same phase and different phase show that the proposed method can both improve the information entropy and the image details evaluation standard comparing with the interpolation method, traditional TV algorithm and MAP algorithm, which indicate that our method can obviously highlight image details and contains more ground texture information. A large number of experiment results reveal that the proposed method is robust and universal for different kinds of ZY-3 satellite images.

show abstract

“…So to convert the ill posed problem into good problem a regularization approach is used. These methods adopt a Bayesian approach [9], according to which the information that can be extracted from LR images about the unknown signal HR image is contained in the probability function of the unknown. The usual methods of regularisation are HMRF, Tikhonov and Total Variation(TV) [10] and the main problem is selection of regularization coefficient .In the above methods the regularization coefficient is constant so the reconstruction result is ideal and the iterative size is still certain real number this leads to poor reconstruction to overcome that in proposed method( a new hybrid method combined both SR and PSO algorithm) the iteration step size is depend on the fitness value [6], when it is reached minimum the estimated super resolution image/video is optimised.PSO [8] as an efficient optimization method is easy to implement, Has strong global convergence ability, robustness and suitable for solving optimization functions in complex environment.…”

Section: Introductionmentioning

confidence: 99%

Image/Video Reconstruction Using a Novel Hybrid Method

Dasu¹

2017

IJRASET

View full text Add to dashboard Cite

Image/Video super-resolution is an essential part for various technologies, including Video Surveillance, Robotics, Medical applications and Multimedia. Aiming at improve the super resolution of an image/video reconstruction a novel hybrid method combining both super resolution and particle swarm optimization algorithm is proposed. In the process of image/video reconstruction, initially to up-sample the frames a non uniform interpolation method is applied but the frames are still blurry. So to estimate the blur a adaptive regularization approach is used, it consists of fidelity and regularization terms and they are updated by adaptive iteration process .To preserve the edges and remove the noise a Relaxed median filter is used it performs well at any type of noise. Then the reconstructed frames are optimised using the particle swarm optimization algorithm which includes particle input, particle position, motion equations and fitness function .Image/Video reconstruction using existing methods is poor due to fixed iteration step size. To overcome that limitation, in proposed method the iteration step size is adaptively selected based on the fitness value, when it is reached minimum the estimated super resolution image/video is optimised. The quality factors like PSNR, NAE, IEF, Correlation Coefficient and Structural Content are very much improved than existing methods.

show abstract

A Bayesian approach to adaptive video super resolution

Cited by 236 publications

References 27 publications

Real-Time Video Scaling Based on Convolution Neural Network Architecture

Real-Time Video Scaling Based on Convolution Neural Network Architecture

Super Resolution Reconstruction Based on Adaptive Detail Enhancement for Zy-3 Satellite Images

Image/Video Reconstruction Using a Novel Hybrid Method

Contact Info

Product

Resources

About