Neural Network-Based Error Concealment For VVC

In this paper, an error concealment scheme for neural-network based compression of depth image in 3D videos is proposed. In the neural-network based compression, each depth image is represented by one or more neural networks. The advantage of neural-network based compression lies in the parallel processing ability of multiple neurons, which can handle the massive data volume of 3D videos. The similarity of neuron weights of neighboring nodes is exploited to recover the loss neuron weights when transmitting with an error-prone communication channel. With a simulated noisy channel, the quality of compressed 3D video, which is reconstructed undergoing the noisy channel, can be recovered well by the proposed error concealment scheme.

Download Full-text

Image Error Concealment Based on Deep Neural Network

Algorithms ◽

10.3390/a12040082 ◽

2019 ◽

Vol 12 (4) ◽

pp. 82

Author(s):

Zhiqiang Zhang ◽

Rong Huang ◽

Fang Han ◽

Zhijie Wang

Keyword(s):

Neural Network ◽

Deep Neural Network ◽

Error Concealment ◽

Image Features ◽

Local Information ◽

Local Correlation ◽

Spatial Image ◽

Support Area ◽

Non Local ◽

Perceptual Metrics

In this paper, we propose a novel spatial image error concealment (EC) method based on deep neural network. Considering that the natural images have local correlation and non-local self-similarity, we use the local information to predict the missing pixels and the non-local information to correct the predictions. The deep neural network we utilize can be divided into two parts: the prediction part and the auto-encoder (AE) part. The first part utilizes the local correlation among pixels to predict the missing ones. The second part extracts image features, which are used to collect similar samples from the whole image. In addition, a novel adaptive scan order based on the joint credibility of the support area and reconstruction is also proposed to alleviate the error propagation problem. The experimental results show that the proposed method can reconstruct corrupted images effectively and outperform the compared state-of-the-art methods in terms of objective and perceptual metrics.

Download Full-text

A novel error concealment approach based on general regression neural network

2011 International Conference on Consumer Electronics, Communications and Networks (CECNet) ◽

10.1109/cecnet.2011.5768232 ◽

2011 ◽

Cited By ~ 1

Author(s):

Shih-Chun Shao ◽

Jun-Horng Chen

Keyword(s):

Neural Network ◽

Error Concealment ◽

General Regression Neural Network ◽

General Regression

Download Full-text

An Efficient Generalized Error Concealment in Video Codec

International Journal of Computer Vision and Image Processing ◽

10.4018/ijcvip.2020100101 ◽

2020 ◽

Vol 10 (4) ◽

pp. 1-28

Author(s):

Ansari Vaqar Ahmed ◽

Uday Pandit Khot

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Mahalanobis Distance ◽

Euclidean Distance ◽

Error Concealment ◽

Motion Vector ◽

Fuzzy Reasoning ◽

Video Codec ◽

Compression Scheme ◽

Recovery Method

Efficient error concealment (EC) predictor can recover more significant features or structures of entire lost MBs using a pre-transmission algorithm (PTA) with convolutional neural network (CNN) and fuzzy reasoning to select appropriate EC for reconstruction in generalized video-codec compression scheme such as H.264/H.265, etc. Here, the pixel-based motion vector with partition (PMVP) algorithm is modified by using Mahalanobis distance (MD) rather than Euclidean distance (ED) for better MVs recovery. This modified pixel-based motion vector with partition (MPMVP) algorithm is upgraded by two different strategies: one by using voting priority of available MVs based on the probabilities of similar directions and the second by considering separate horizontal and vertical directions of available MVs in voting priority. Similarly, a modified spiral pixel reconstruction (MSPR) algorithm based on directional edge recovery method using minimum and maximum MD from available pixels of surrounding MBs is proposed. The proposed PTA-based modified ECs gives 20.4%, 3.47%, and 6.66% increase in PSNR.

Download Full-text