Spatial Error Concealment With an Adaptive Linear Predictor

Liu, Jing; Zhai, Guangtao; Yang, Xiaokang; Yang, Bing; Chen, Li

doi:10.1109/tcsvt.2014.2359145

Cited by 42 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When original pixel is not available (e.g. missed during transmission), the proposed algorithm can proceed to fill in the missing part with preferable visual quality [30]. In contrast, the procedure of MRP does not work as there is no explicit underlying image model.…”

Section: B Comparison With Mrpmentioning

confidence: 99%

Lossless Predictive Coding for Images With Bayesian Treatment

Liu

Zhai

Yang

et al. 2014

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Adaptive predictor has long been used for lossless predictive coding of images. Most of existing lossless predictive coding techniques mainly focus on suitability of prediction model for training set with the underlying assumption of local consistency, which may not hold well on object boundaries and cause large predictive error. In this paper, we propose a novel approach based on the assumption that local consistency and patch redundancy exist simultaneously in natural images. We derive a family of linear models and design a new algorithm to automatically select one suitable model for prediction. From the Bayesian perspective, the model with maximum posterior probability is considered as the best. Two types of model evidence are included in our algorithm. One is traditional training evidence, which represents the models’ suitability for current pixel under the assumption of local consistency. The other is target evidence, which is proposed to express the preference for different models from the perspective of patch redundancy. It is shown that the fusion of training evidence and target evidence jointly exploits the benefits of local consistency and patch redundancy. As a result, our proposed predictor is more suitable for natural images with textures and object boundaries. Comprehensive experiments demonstrate that the proposed predictor achieves higher efficiency compared with the state-of-the-art lossless predictors.

show abstract

Section: B Comparison With Mrpmentioning

confidence: 99%

Lossless Predictive Coding for Images With Bayesian Treatment

Liu

Zhai

Yang

et al. 2014

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

show abstract

“…In 2009, Thuong Le-Tien and colleagues [7] used a SISR to recreate super-resolution images by fusing the frequency domain with the wavelet domain. On the other hand, in terms of SR image reconstruction, the Deep Learning technique has been investigated and put into practise, which has resulted in some exciting discoveries [8]. introduced the SRCNN in an effort to improve the upscaling of low-resolution (LR) images to high-resolution (HR) images, and they reported positive results; however, they also acknowledged that accelerating deep models and having an indepth understanding of deep models, as well as the criteria for designing and evaluating the objective functions, are challenges for optimising their model [9].…”

Section: Introductionmentioning

confidence: 99%

Advancements in Image Super-Resolution: A Deep Learning Approach

Venkatesan

2023

IJSMIEN

View full text Add to dashboard Cite

The goal of this study is to address the problem of picture super-resolution (SR) by using a deep learning strategy to the opposite problem. Super-resolution (SR) imaging has garnered attention for a long time. Since then, several different branches of machine learning and deep learning have made significant strides in solving problems like these in medical imaging. Both deep learning and machine learning belong to this category. We explore the many applications of the Generative Adversarial Network (GAN) method, and talk about the deep learning strategies that may be utilised to address the problem of picture super-resolution.The Enhanced Super-Resolution Generative Adversarial Network (ESRGAN) and the Residual in Residual Dense Network (RRDN) are two state-of-the-art systems that have been created for picture super-resolution by the 'idealo' group. In this study, we analyse the effectiveness of both of these networks. The ESRGAN and RRDN teams are responsible for the development of both of these methodologies. The proposed model will be retrained with new parameters, and its results will be compared to those of two state-of-the-art techniques, which are adversarial networks for super-resolution of a single picture and enhanced super-resolution adversarial networks. Both of these approaches are aimed at improving the resolution of a single image. It can be good to understand the specifics of the required model as well as how our parameter choices compare to those of the other team in this particular circumstance.

show abstract

“…The proposed technique comprises block classification, edge direction interpolation, and filtering interpolation. A novel spatial error concealment algorithm with an adaptive linear predictor is proposed in [10]. Under the sequential recovery framework, pixels in missing blocks are successively reconstructed based on adaptive linear predictor.…”

Section: Introductionmentioning

confidence: 99%

Disparity-based Error Concealment for Stereoscopic Images with Superpixel Segment

2018

KSII TIIS

View full text Add to dashboard Cite

To solve the problem of transmission errors in stereoscopic images, this paper proposes a novel error concealment (EC) method using superpixel segmentation and adaptive disparity selection (SSADS). Our algorithm consists of two steps. The first step is disparity estimation for each pixel in a reference image. In this step, the numbers of superpixel segmentation labels of stereoscopic images are used as a new constraint for disparity matching to reduce the effect of mismatching. The second step is disparity selection for a lost block. In this step, a strategy based on boundary smoothness is proposed to adaptively select the optimal disparity which is used for error concealment. Experimental results demonstrate that compared with other methods, the proposed method has significant advantages in both objective and subjective quality assessment.

show abstract

Spatial Error Concealment With an Adaptive Linear Predictor

Cited by 42 publications

References 42 publications

Lossless Predictive Coding for Images With Bayesian Treatment

Lossless Predictive Coding for Images With Bayesian Treatment

Advancements in Image Super-Resolution: A Deep Learning Approach

Disparity-based Error Concealment for Stereoscopic Images with Superpixel Segment

Contact Info

Product

Resources

About