Compression of depth maps is important for “texture plus depth” format of multiview images, which enables synthesis of novel intermediate views via depth-image-based rendering (DIBR) at decoder. Previous depth map c...
详细信息
Compression of depth maps is important for “texture plus depth” format of multiview images, which enables synthesis of novel intermediate views via depth-image-based rendering (DIBR) at decoder. Previous depth map coding schemes exploit unique depth data characteristics to compactly and faithfully reproduce the original signal. In contrast, since depth map is only a means to the end of view synthesis and not itself viewed, in this paper we explicitly manipulate depth values, without causing severe synthesized view distortion, in order to maximize representation sparsity in the transform domain for compression gain - we call this process transform domain spar-sification (TDS). Specifically, for each pixel in the depth map, we first define a quadratic penalty function, with minimum at ground truth depth value, based on synthesized view's distortion sensitivity to the pixel's depth value during DIBR. We then define an objective for a depth signal in a block as a weighted sum of: i) signal's sparsity in the transform domain, and ii) per-pixel synthesized view distortion penalties for the chosen signal. Given that sparsity (l 0 -norm) is non-convex and difficult to optimize, we replace the l 0 -norm in the objective with a computationally inexpensive weighted l 2 -norm; the optimization is then an unconstrained quadratic program, solvable via a set of linear equations. For the weighted l 2 -norm to promote sparsity, we solve the optimization iteratively, where at each iteration weights are readjusted to mimic sparsity-promoting l τ -norm, 0 ≤ τ ≤ 1. Using JPEG as an example transform codec, we show that our TDS approach gained up to 1.7dB in rate-distortion performance for the interpolated view over compression of unaltered depth maps.
We propose a natural scene statistic based Blind/Referenceless Image Spatial QUality Evaluator (BRISQUE) which extracts the point wise statistics of local normalized luminance signals and measures image naturalness (o...
详细信息
We propose a natural scene statistic based Blind/Referenceless Image Spatial QUality Evaluator (BRISQUE) which extracts the point wise statistics of local normalized luminance signals and measures image naturalness (or lack there of) based on measured deviations from a natural image model. We also model the distribution of pairwise statistics of adjacent normalized luminance signals which provides distortion orientation information. Although multi scale, the model uses easy to compute features making it computationally fast and time efficient. The frame work is shown to perform statistically better than other proposed no reference algorithms and full reference structural similarity index (SSIM).
In this paper, we present a forensic algorithm to discriminate between original and forged regions in JPEG images, under the hypothesis that the tampered image presents a non-aligned double JPEG compression (NA-JPEG)....
详细信息
In this paper, we present a forensic algorithm to discriminate between original and forged regions in JPEG images, under the hypothesis that the tampered image presents a non-aligned double JPEG compression (NA-JPEG). Unlike previous approaches, the proposed algorithm does not need to manually select a suspect region to test the presence or the absence of NA-JPEG artifacts. Based on a new statistical model, the probability for each 8 × 8 DCT block to be forged is automatically derived. Experimental results, considering different forensic scenarios, demonstrate the validity of the proposed approach.
Frame-compatible format, packing two neighbouring views into one frame, is considered as a promising solution for 3D distribution in the existing system. In this paper, a fast Intra encoder is designed for frame-compa...
详细信息
Frame-compatible format, packing two neighbouring views into one frame, is considered as a promising solution for 3D distribution in the existing system. In this paper, a fast Intra encoder is designed for frame-compatible format coding based on the content similarity and proposed to reduce the computation complexity. With a relative shift, statistics analysis proves the high prediction correlation between the two packed views and qualifies the first coded view provide prediction mode as the reference for the second view prediction. The proposed scheme enables the prediction of the second view's MB only perform candidate Intra modes and candidate directions of each, according to the reference MB in the first view. An average of 75.69% complexity reduction of encoding the second view can be achieved with comparable compression efficiency. With no requirement to change the decoder of existing system and there is little effort for this encouraging format, this proposed algorithm improves the efficiency of frame-compatible format coding, and contributes to the delivery of 3D service.
This paper proposes an S-model to normalize the parameters measured from MPEG TS [1][2]. Using the additive normalized scores, an overall assessment score of the quality customer perceived is derived. This paper also ...
详细信息
This paper proposes an S-model to normalize the parameters measured from MPEG TS [1][2]. Using the additive normalized scores, an overall assessment score of the quality customer perceived is derived. This paper also provides a systematic approach to calculate the coefficients used in this model. This model has been applied to the actual operational environment, not only to detect the streaming quality, but also to provide quality trend analysis and degradation area judgment.
In this paper a robust digital image watermarking algorithm proposed in wavelet transform domain. At first, the original image is decomposed into ten sub-bands by the 3-level wavelet transform. Then, the values of wav...
详细信息
In this paper a robust digital image watermarking algorithm proposed in wavelet transform domain. At first, the original image is decomposed into ten sub-bands by the 3-level wavelet transform. Then, the values of wavelet coefficients in HL3 are quantized. For embedding watermark bits, trees of wavelet coefficients are used. In each tree, the distance between the average of four wavelet coefficients in HL2 and their corresponding HL3 coefficient determines how HL3 coefficient is quantized. The experimental results show that the proposed algorithm has a good resistance against to common attacks, especially JPEG compression; moreover it outperforms previous methods in the most situations.
We present techniques for processing MPEG-audio encoded signals during the decoding process, using efficient fixed-point arithmetic operations. A large signal-to-quantization-noise-ratio is achieved over a large range...
详细信息
We present techniques for processing MPEG-audio encoded signals during the decoding process, using efficient fixed-point arithmetic operations. A large signal-to-quantization-noise-ratio is achieved over a large range of input levels. By taking advantage of MPEG-audio built-in properties, quantization distortion at the outputs of our systems is kept largely inaudible, even though only low-resolution fixed-point operations are used in the processing.
This work presents the splitting of the JPEG2000 file into two parts to be stored on two different systems. One fragment, which contains the most important information, is stored in a system with a high reliability bu...
详细信息
This work presents the splitting of the JPEG2000 file into two parts to be stored on two different systems. One fragment, which contains the most important information, is stored in a system with a high reliability but slow transfer rate and small capacity. The other fragment is stored in a high-capacity and fast speed but lower reliability system.
On the in Internet today, there are numerous web-sites offering places to store and share images (photos, drawings, cliparts, etc.). Online Social Networks (OSN) and Online Photo Services (OPS) usually offer this type...
详细信息
On the in Internet today, there are numerous web-sites offering places to store and share images (photos, drawings, cliparts, etc.). Online Social Networks (OSN) and Online Photo Services (OPS) usually offer this type of service making it possible to upload and manage the images they publish. Users can publish their own images as well as add descriptions and tags. By adjusting the personal privacy settings, the images become visible only to authorized users, with it then being possible to add any comments. Generally, OSN/OPS manipulate the published images by resizing, upgrading metadata, compressing, embedding watermarking, and making it difficult to use well-known steganographic methodologies on them. In this paper two new steganographic schemes that mainly take advantage of the published images and survive the OSN/OPS processing before their publication, are proposed and analyzed. Finally, a notification system which uses tags to manage the sent/received information on the secret channel is proposed. This allows for the deletion of the received information as well as a bandwidth increase of the secret channel.
We implemented and evaluated a heterogeneous system architecture that combines a traditional computer cluster with a broadband network of embedded set-top box (STB) devices to provide a distributed computing platform ...
详细信息
We implemented and evaluated a heterogeneous system architecture that combines a traditional computer cluster with a broadband network of embedded set-top box (STB) devices to provide a distributed computing platform for parallel applications. Our prototype system for broadband grid computing leverages the recent dramatic progress in computational power of STBs. It includes a complete head-end cable system based on the Tru2way standard, a DOCSIS-2.0 network, and an implementation of the Open MPI library running on the STB embedded operating system across 128 devices. An important contribution of our work is a novel method for the virtualization of a large collection of embedded processors within a managed broadband network. This enables the embedded processors to transparently inter-operate with servers in the computer cluster using the message-passing model. To evaluate the interoperability, performance, and scalability of our system we completed a set of experiments with the standard IMB MPI benchmark suite as well as two real parallel applications. The experimental results confirm that there is an important convergence trend between traditional computing and embedded computing and that a broadband network of embedded processors is a promising new platform for a variety of computationally-intensive and data-intensive grid applications.
暂无评论