WynerZiv video coding for wireless lightweight multimedia applications
 Nikos Deligiannis^{1, 2}Email author,
 Frederik Verbist^{1, 2},
 Athanassios C Iossifides^{3},
 Jürgen Slowack^{2, 4},
 Rik Van de Walle^{2, 4},
 Peter Schelkens^{1, 2} and
 Adrian Munteanu^{1, 2}
DOI: 10.1186/168714992012106
© Deligiannis et al; licensee Springer. 2012
Received: 10 October 2011
Accepted: 14 March 2012
Published: 14 March 2012
Abstract
Wireless video communications promote promising opportunities involving commercial applications on a grand scale as well as highly specialized niche markets. In this regard, the design of efficient video coding systems, meeting such key requirements as low power, mobility and low complexity, is a challenging problem. The solution can be found in fundamental information theoretic results, which gave rise to the distributed video coding (DVC) paradigm, under which lightweight video encoding schemes can be engineered. This article presents a new hashbased DVC architecture incorporating a novel motioncompensated multihypothesis prediction technique. The presented method is able to adapt to the regional variations in temporal correlation in a frame. The proposed codec enables scalable WynerZiv video coding and provides stateoftheart distributed video compression performance. The key novelty of this article is the expansion of the application domain of DVC from conventional video material to medical imaging. Wireless capsule endoscopy in particular, which is essentially wireless video recording in a pill, is proven to be an important application field. The low complexity encoding characteristics, the ability of the novel motioncompensated multihypothesis prediction technique to adapt to regional degrees of temporal correlation (which is of crucial importance in the context of endoscopic video content), and the high compression performance make the proposed distributed video codec a strong candidate for future lightweight (medical) imaging applications.
Keywords
WynerZiv coding distributed video coding hashbased motion estimation wireless lightweight multimedia applications1. Introduction
Traditional video coding architectures, like the H.26x [1] recommendations, mainly target broadcast applications, where video content is distributed to multiple users, and focus on optimizing the compression performance. The source redundancy is exploited at the encoder by means of predictive coding. In this way, traditional video coding implies joint encoding and decoding of video. Namely, the encoder produces a prediction of the source and then codes the difference between the source and its prediction. Motioncompensated prediction in particular, a key algorithm to achieve high compression performance by removing the temporal correlation between successive frames in a sequence, is very effective but computationally demanding.
The need for highly efficient video compression architectures maintaining lightweight encoding remains challenging in the context of wireless video capturing devices that have only modest computational capacity or operate on limited battery life. The solution to reduce the encoding complexity can be found in the fundamentals of information theory, which constitute an original coding perspective, known as distributed source coding (DSC). The latter stems from the theory of Slepian and Wolf [2] on lossless separate encoding and joint decoding of correlated sources. Subsequently, Wyner and Ziv [3] extended the DSC problem to the lossy case, deriving the rate distortion function with side information at the decoder. Driven by these principles, the distributed, alias WynerZiv, video coding paradigm has arisen [4, 5].
Unlike traditional video coding, in distributed video coding (DVC), the source redundancies are exploited at the decoder side, implying separate encoding and joint decoding. Specifically, a prediction of the source, named side information, is generated at the decoder by using the already decoded information. By expressing the statistical dependency between the source and the side information in the form of a virtual correlation channel, e.g. [4–8], compression can be achieved by transmitting parity or syndrome bits of a channel code, which are used to decode the source with the aid of the side information. Hence, computationally expensive tasks, like motion estimation, could be relocated to the decoder, allowing for a flexible sharing of the computational complexity between the encoder and the decoder and enabling the design of lightweight encoding architectures.
DVC has been recognized as a potential strategic component for a wide range of lightweight video encoding applications, including visual sensor networks and wireless lowpower surveillance [9, 10]. A unique application of particular interest in this article is wireless capsule endoscopy^{a}. Conventional endoscopy, like colonoscopy or gastroscopy, has proven to be an indispensable tool in the diagnosis and remedy of various diseases of the digestive track. Significant advances in miniaturization have led to the emergence of wireless capsule endoscopy [11]. At the size of a large pill, a wireless capsule endoscope comprises a light source, an integrated chip video camera, a radio telemetry transmitter and a limited lifespan battery. The smallscale nature of the recording device forces severe constraints on the required video coding technology, in terms of computational complexity, operating time, and power consumption. Moreover, since the recorded video is used for medical diagnosis, highquality decoded video at an efficient compression ratio is of paramount importance.
Generating highquality side information plays a vital role in the compression performance of a DVC system. Conversely to traditional predictive coding, in DVC the original frame is not available during motion estimation, since this is performed at the decoder. Producing accurate motioncompensated predictions at the decoder for a wide range of video content, while at the same time constraining the encoding complexity and guaranteeing high compression performance, poses a major challenge. This problem becomes even more intricate in the largely unexplored application of DVC in wireless capsule endoscopy, in which the recorded video material contains extremely irregular motion, due to low frame acquisition rates and the erratic movement of the capsule along the gastrointestinal track. Towards tackling this challenge, this study presents a novel hashbased DVC architecture.
First and foremost, this study paves the road for the application of DVC systems in lightweight medical imaging where the proposed codec achieves high compression efficiency with the additional benefit of low computational encoding complexity. Second, the proposed WynerZiv video codec incorporates a novel motioncompensated multihypothesis prediction scheme, that supports online tuning to the spatial variations in temporal correlation in a frame by obtaining information from the coded hash in case temporal prediction is unreliable. Third, this article includes a thorough experimental evaluation of the proposed hashbased DVC scheme on (i) conventional test sequences, numerous (ii) traditional endoscopic as well as (iii) wireless capsule endoscopic video content. The experimental results show that the proposed DVC outperforms alternative DVC schemes, including DISCOVER, the hashbased DVC from [12] and our previous study [13], as well as conventional codecs, namely, Motion JPEG and H.264/AVC Intra [1]. Four, this article incorporates a detailed analysis of the encoding complexity and buffer size requirements of the proposed system.
The rest of the article is structured as follows. Section 2 covers an overview of SlepianWolf and WynerZiv coding and their instantiation in DVC. Section 3 describes two application scenarios, both relevant to DVC in general and the proposed video codec in particular. Our novel DVC codec is explained in Section 4 and experimentally evaluated in Section 5, using conventional test sequences as well as endoscopic test video. Section 6 draws the conclusions of this study.
2. Background and contributions
2.1. SlepianWolf coding
Consider the compression of two correlated, discrete, identically and independently distributed (i.i.d.) random sources X and Y. According to Shannon's source coding theory [14], the achievable lower rate bound for lossless joint encoding and decoding is given by the joint entropy H(X, Y) of the sources. Slepian and Wolf [2] studied the lossless compression scenario in which the sources are independently encoded and jointly decoded. According to their theory, the achievable rate region for decoding X and Y with an arbitrarily small error probability is given by R_{ X } ≥ H(XY), R_{ Y } ≥ H(YX), R_{ X } + R_{ Y } ≥ H(X, Y), where H(XY) and H(YX) are the conditional entropies of the considered sources, and R_{ X }, R_{ Y } are the respective rates at which the sources X and Y are coded, i.e., the SlepianWolf theorem states that even when correlated sources are encoded independently, a total rate close to the joint entropy suffices to achieve lossless compression.
The SlepianWolf theorem constructs a random binning argument, in which the employed code generation is asymptotic and nonconstructive. In [15], Wyner pointed out the strong relation between random binning and channel coding, suggesting the use of linear channel codes as a practical solution for SlepianWolf coding. Wyner's methodology was recently used by Pradhan and Ramchandran [16], in the context of practical SlepianWolf code design based on conventional channel codes like block and trellis codes. In the particular case of binary symmetric correlation between the sources, Wyner's scheme can be extended to stateoftheart binary linear codes, such as Turbo [5, 17], and lowdensity paritycheck (LDPC) codes [18], approaching the SlepianWolf limit. A turbo scheme with structured component codes was used in [17] while parity bits instead of syndrome bits were sent in [5]. Although breaking the close link with channel coding, characterized by syndromes and coset codes, the latter solutions offer inherent robustness against transmission errors.
2.2. WynerZiv coding
WynerZiv coding [3] refers to the problem of lossy compression with decoder side information. Suppose X and Y are two statistically dependent i.i.d. random sources, where X is independently encoded and decoded using Y as side information. The reconstructed source $\widehat{X}$ has an expected distortion $D=Ed\left(x,\widehat{x}\right)$. According to the WynerZiv theorem [3], a rate loss is sustained when the encoder is ignorant of the side information, namely ${R}_{X\leftY\right.}^{*}\left(D\right)\ge {R}_{X\leftY\right.}\left(D\right)$, where ${R}_{X\leftY\right.}^{*}\left(D\right)$ is the WynerZiv rate and R_{ XY }(D) is the rate when the side information is available to the encoder as well. However, Wyner and Ziv further showed that equality holds for the quadratic Gaussian case, namely the case where X and Y are jointly Gaussian and a meansquare distortion metric d(•,•) is used.
Initial practical WynerZiv code design focused on finding good nested codes among lattice [19] and trellisbased codes [16] for the quadratic Gaussian case. However, as the dimensionality increases, lattice source codes approach the source coding limit much faster than lattice channel codes approach capacity. This observation has induced the second wave of WynerZiv code design which is based on nested lattice codes followed by binning [20]. The third practical approach to WynerZiv coding considers nonnested quantization followed by efficient binning, realized by a highdimensional channel code [5]. Other constructions in the literature propose turbotrellis WynerZiv codes, in which trellis coded quantization is concatenated with a Turbo [21] or an LDPC [22] code.
2.3. DVC
One of the applications of DSC that has received a substantial amount of research attention is DVC. Except for providing lowcomplexity encoding solutions for video, WynerZiv coding has been shown to provide error resilient video coding by means of distributed jointsource channel coding [23], or systematic forward error protection [24]. Moreover, layered WynerZiv code [25] constructions support scalable video coding [23].
An early practical DVC implementation was the PRISM codec [4], combining BoseChaudhuriHocquenghem channel codes with efficient entropy coding and performing blockbased joint decoding and motion estimation. An additional CRC check was sent to the decoder to select between many decoded versions of a block, each version in fact corresponding to a different motion vector. An alternative DVC architecture, that implemented WynerZiv coding as quantization followed by turbo coding using a feedback channel to enable decoderdriven optimal rate control, was presented in [5]. In this architecture, side information was generated at the decoder using motioncompensated interpolation (MCI). The architecture was further improved upon, resulting in the DISCOVER codec [26], which included superior MCI [27] through blockbased bidirectional motion estimation and compensation combined with spatial smoothing. The DISCOVER codec is a wellestablished reference in DVC, delivering stateoftheart compression performance.
In sequences with highly irregular motion content, blind motion estimation at the decoder, by means of MCI for example, fails to deliver adequate prediction quality. One technique to overcome this problem is to perform hashbased motion estimation at the decoder. Aaron et al. [28] proposed a hash code consisting of a coarsely subsampled and quantized version of each block in a WynerZiv frame. The encoder performed a blockbased decision whether to transmit the hash. For the blocks for which a hash code was sent, hashbased motion estimation was carried out at the decoder, while for the rest of the blocks, for which no hash was sent, the colocated block in the previous reconstructed frame was used as side information. In [29], several hash generation approacheseither in the pixel or in the transform domainwere investigated. It was shown that hash information formed by a quantized selection of lowfrequency DCT bands per block was outperforming the other methods [29]. In [12], a blockbased selection, based on the current frame to be coded and its future and past frames in hierarchical order, was performed at the encoder. Blocks for which MCI was foreseen to fail were lowquality H.264/AVC Intra encoded and transmitted to the decoder to assist MCI. The residual frame, given by the difference between all reconstructed intra coded blocks or the central luminance value (for nonhash blocks) and the corresponding blocks in the WynerZiv frame, was formed and WynerZiv encoded. In our previous study [30], we have introduced a hashbased DVC, where the auxiliary information conveyed to the decoder comprised a number of most significant bitplanes of the original WynerZiv frames. Such a bitplanebased hash facilitates accurate decoderside motion estimation and advanced probabilistic motion compensation [31]. Transformdomain WynerZiv encoding was applied on the remaining least significant bitplanes, defined as the difference of the original frame and the hash [31]. In [32], hashbased motion estimation was combined with side information refinement to further improve the compression performance at the expense of minimal structural decoding delay.
Driven by the requirements of niche applications like wireless capsule endoscopy, this study proposes a novel hashbased DVC architecture introducing the following novelties. First, in contrast to our previous DVC architectures [30, 31], which employed a bitplane hash, the presented system generates the hash as a downscaled and subsequently conventionally intra coded version of the original frames. Second, unlike our previous study [30–32], the hash is exploited in the design of a novel motioncompensated multihypothesis prediction scheme, which is able to adapt to the regional variations in temporal correlation in a frame by extracting information from the hash when temporal prediction is untrustworthy. Compared to alternative techniques in the literature, i.e., [12, 13, 26, 27], the proposed methodology delivers superior performance under strenuous conditions, namely, when irregular motion content is encountered as in for example endoscopic video material, where gastrointestinal contractions can generate severe morphological distortions in conjunction with extreme camera panning. Third, the way the hash is constructed and utilized to generate side information in the proposed codec also differs from the approaches in [28, 29]. Fourth, conversely to alternative hashbased DVC systems [12, 31], the proposed architecture codes the entire frames using powerful channel codes instead of coding only the difference between the original frames and the hash. Fifth, unlike existing works in the literature, this article experimentally shows the stateoftheart compression performance of the proposed DVC not only on conventional test sequences, but also on traditional and wireless capsule endoscopic video content, while lowcost encoding is guaranteed.
3. Application scenarios for DVC
3.1. Wireless lightweight manytomany video communication
WynerZiv video coding can be a key component to realize manytomany video streaming over wireless networks. Such a setting demands optimal video streams, tailored to specific requirements in terms of quality, framerate, resolution, and computational capabilities imposed by a set of recorders and receivers. Consider a network of wireless visual sensors that is deployed to monitor specific scenes, providing security and surveillance. The acquired information is gathered by a central node for decoding and processing. Wireless network surveillance applications are characterized by a wide variety of scene content, ranging from complex motion sequences, e.g., crowd or traffic monitoring, to surveillance of scenes mostly devoid of significant motion, e.g., fire and home monitoring.
In such scenarios, wireless visual sensors are understood to be cheap, battery powered and modest in terms of complexity. In this concept, WynerZiv video coding facilitates communications from the sensors to the central base station, by maintaining low computational requirements at the recording sensor, while simultaneously ensuring fast, highly efficient, and scalable coding. From a complementary perspective, a conventional predictive video coding format with lowcomplexity decoding characteristics provides a broadcast oriented onetomany video stream for further dissemination from the base station. Such a video communications' scenario centralizes the computational complexity in the fixed network infrastructure, which would be responsible for transcoding the WynerZiv video coding streams to a conventional format.
3.2. Wireless capsule endoscopy
Focussing on the video coding technology part, it is apparent that wireless endoscopy is subjected to severe constraints in terms of available computational capacity and power consumption. Contemporary capsule video chips employ conventional coding schemes operating in a lowcomplexity, intraframe mode, i.e., Motion JPEG [34], or even no compression at all. Current capsule endoscopic video systems operate at modest frame resolutions, e.g., 256 × 256 pixels, and frame rates, e.g., 25 Hz, on a battery life time of approximately 7 h. Future generations of capsule endoscopes are intended to transmit at increased resolution, frame rate, and battery life time and will therefore require efficient video compression at a computational cost as low as possible. In addition, a video coding solution supporting temporal scalability has an attractive edge, enabling increased focus during the relevant stages of the capsules bodily journey. DVC is a strong candidate to fulfil the technical demands imposed by wireless capsule endoscopy, offering lowcost encoding, scalability, and high compression efficiency [10].
4. Proposed DVC architecture
4.1. The encoder
Every incoming frame is categorized as a key or a WynerZiv frame, denoted by K and W, respectively, as to construct groups of pictures (GOP) of the form KW ...W. The key frames are coded separately using a conventional intra codec, e.g., H.264/AVC intra [1] or Motion JPEG.^{b} The WynerZiv frames on the other hand are encoded in two stages. For every WynerZiv frame, the encoder first generates and codes a hash, which will assist the decoder during the motion estimation process. In the second stage, every WynerZiv frame undergoes a discrete cosine transform (DCT) and is subsequently coded in the transform domain using powerful channel codes, thus generating a WynerZiv bit stream.
4.1.1. Hash formation and coding
Our WynerZiv video encoder creates an efficient hash that consists of a lowquality version of the downsized original WynerZiv frames. In contrast to our previous hashbased DVC architectures [30, 31], where the dimensions of the hash were equal to the dimensions of the original input frames, coding a hashbased on the downsampled WynerZiv frames reduces the computational complexity. In particular, every WynerZiv frame undergoes a downscaling operation by a factor, d ∈ ℤ_{+}. To limit the involved operations, straightforward downsampling is applied. Foregoing a lowpass filter to bandlimit, the signal prior to downsampling runs the risk of introducing undesirable aliasing artefacts. However, experimental experience has shown that the impact on the overall ratedistortion (RD) performance of the entire system does not outweigh the computational complexity incurred by the use of stateoftheart downsampling filters, e.g., Lanczos filers [35].
After the dimensions of the original WynerZiv frames have been reduced, the result is coded using a conventional intra video codec, exploiting spatial correlation within the hash frame only. The quality at which the hash is coded has experimentally been selected and constitutes a tradeoff between (i) obtaining a constant quality of the decoded frames, which is of particular interest in medical applications, (ii) achieving high RD performance for the proposed system and (iii) maintaining a low hash rate overhead. We notice that constraining the hash overhead comes with the additional benefit of minimizing the hash encoding complexity. On the other hand, ensuring sufficient hash quality so that the accuracy of the hashbased motion estimation at the decoder is not compromised or so that even pixels in the hash itself could serve as predictors is important. Afterwards, the resulting hash bit stream is multiplexed with the key frame bit stream and sent to the decoder.
We wish to highlight that, apart from assisting motion estimation at the decoder as in contemporary hashbased systems, the proposed hash code is designed to also act as a candidate predictor for pixels for which the temporal correlation is low. This feature is of particular significance especially when difficulttocapture endoscopic video content is coded. To this end, the presented hash generation approach was chosen over existing methods in which the hash consists of a number of most significant WynerZiv frame bitplanes [30, 31], of coarsely subsampled and quantized versions of blocks [28], or of quantized low frequency DCT bands [29] in the WynerZiv frames.
Furthermore, we note that, in contrast to other hashbased DVC solutions [12, 28], the proposed architecture avoids blockbased decisions on the transmission of the hash at the encoder side. Although this can increase the hash rate overhead when easytopredict motion content is coded, it comes at the benefit of constraining the encoding complexity, in the sense that the encoder is not burdened by expensive blockbased comparisons or memory requirements necessary for such mode decision. An additional key advantage of the presented hash code is that it facilitates accurate side information creation using pixelbased multihypothesis compensation at the decoder, as explained in Section 4.2.2. In this way, the presented hash code enhances the RD performance of the proposed system especially for irregular motion content, e.g., endoscopic video material.
4.1.2. WynerZiv encoding
In addition to the coded hash, a WynerZiv layer is created for every WynerZiv frame, providing efficient compression [5] and scalable coding [25]. In line with the DVC architecture introduced in [5], the WynerZiv frames are first transformed with a 4 × 4 integer approximation of the DCT [1] and the obtained coefficients are subsequently assembled in frequency bands. Each DCT band is independently quantized using a collection of predefined quantization matrices (QMs) [26], where the DC and the AC bands are quantized with a uniform and doubledeadzone scalar quantizer, respectively. The quantized symbols are translated into binary codewords and passed to a LDPC Accumulate (LDPCA) encoder [36], assuming the role of SlepianWolf encoder.
The LDPCA [36] encoder realizes Slepian and Wolf's random binning argument [15] through linear channel code syndrome binning. In detail, let b be a binary Mtuple containing a bitplane of a coded DCT band β of a WynerZiv frame, where M is the number of coefficients in the band. To compress b, the encoder employs an (M, k) LDPC channel code C constructed by the generator matrix ${\mathbf{G}}_{k\times M}=\left[{\mathbf{I}}_{k}\phantom{\rule{1em}{0ex}}{\mathbf{P}}_{k\times \left(Mk\right)}\right]$^{c}. The corresponding parity check matrix of C is ${\mathbf{H}}_{\left(Mk\right)\times M}=\left[{\mathbf{P}}_{k\times \left(Mk\right)}^{T}\phantom{\rule{1em}{0ex}}{\mathbf{I}}_{Mk}\right]$. Thereafter, the encoder forms the syndrome vector as s = bH^{ T }. In order to achieve various puncturing rates, the LDPC syndromebased scheme is concatenated with an accumulator [36]. Namely, the derived syndrome bits s are in turn mod2 accumulated, producing the accumulated syndrome tuple α. The encoder stores the accumulated syndrome bits in a buffer and transmits them incrementally upon the decoder's request using a feedback channel, as explained in Section 4.2.3. Note that contemporary wireless (implantable) sensorsincluding capsule endoscopessupport bidirectional communication [33, 37, 38]. That is, a feedback channel from the encoder to the decoder is a viable solution for the pursued applications. The effect of the employed feedback channel on the decoding delay, and in turn on the buffer requirements at the encoder of a wireless capsule endoscope, is studied in Section 5.3.
Note that the focus of this study is to successfully target various lightweight applications by improving the compression efficiency of WynerZiv video coding while maintaining low computational cost at the encoder. Hence, in order to accurately evaluate the impact of the proposed techniques on the RD performance, the proposed system employs LDPCA codes which are also used in the stateoftheart codecs of [13, 26]. Observe that for distributed compression under a noiseless transmission scenario the syndromebased SlepianWolf scheme [15] is optimal since it can achieve the information theoretical bound with the shortest channel codeword length [23]. Nevertheless, in order to address distributed joint sourcechannel coding (DJSCC) in a noisy transmission scenario the paritybased [23] SlepianWolf scheme needs to be deployed. In the latter, paritycheck bits are employed to indicate the SlepianWolf bins, thereby achieving equivalent SlepianWolf compression performance at the cost of an increased codeword length [23].
It is important to mention that, conversely to other hashdriven WynerZiv schemes operating in the transform domain, e.g., [12, 31], the presented WynerZiv encoder encodes the entire original WynerZiv frame, instead of coding the difference between the original frame and the reconstructed hash. The motivation for this decision is twofold. The first reason stems from the nature of the hash. Namely, coding the difference between the WynerZiv frame and the reconstructed hash would require decoding and interpolating the hash at the encoder, an operation which is computationally demanding and would pose an additional strain on the encoder's memory demands. Second, compressing the entire WynerZiv frame with linear channel codes enables the extension of the scheme to the DJSCC case [23], thereby providing errorresilience for the entire WynerZiv frame if a parity based SlepianWolf approach is followed.
4.2. The decoder
The main components of the presented DVC architecture's decoding process are treated separately, namely dealing with the hash, side information generation and WynerZiv decoding. The decoder first conventionally intra decodes the key frame bit stream and stores the reconstructed frame in the reference frame buffer. In the following phase, the hash is handled, which is detailed next.
4.2.1. Hash decoding and reconstruction
The hash bitstream is decoded with the appropriate conventional intra codec. The reconstructed hash is then upscaled to the original WynerZiv frame's resolution. The ideal upscaling process consists of upsampling followed by ideal interpolation filtering. The ideal interpolation filter is a perfect lowpass filter with gain d and cutoff frequency π/d without transition band [39]. However, such a filter corresponds to an infinite length impulse response h_{ideal}, to be precise, a sinc function h_{ideal}(n) = sinc(n/d) where n ∈ ℤ_{+}, which cannot be implemented in practice.
Such interpolation filter is known in the literature as a Lanczos3 filter [35]. Following [40], the resulting filter taps are normalized to obtain unit DC gain while the input samples are preserved by the upscaling process since h_{0}(n) = 1.
4.2.2. Side information generation
After the hash has been restored to the same frame size as the original WynerZiv frames, it is used to perform decoderside motion estimation. The quality of the side information is an important factor on the overall compression performance of any WynerZiv codec, since the higher the quality the less channel code rate is required for WynerZiv decoding. The proposed side information generation algorithm performs bidirectional overlapped block motion estimation (OBME) using the available hash information and a past and a future reconstructed WynerZiv and/or key frame as references.
Temporal prediction is carried out using a hierarchical frame organization, similar to the prediction structures used in [5, 12, 26]. It is important to note that conversely to our previous study [30], in which motion estimation was based on bitplanes, this study follows a different approach regarding the nature of the hash as well as the block matching process. Before motion estimation is initiated, the reference frames are preprocessed. Specifically, to improve the consistency of the resulting motion vectors, the reference frames are first subjected to the same downsampling and interpolation operation as the hash.
where p visits all the colocated pixel positions in the blocks ${\stackrel{\u0303}{W}}_{\mathbf{u}}$ and ${\stackrel{\u0303}{R}}_{k,\mathbf{u}\mathbf{v}}$, respectively. The motion search is executed at integerpel accuracy and the obtained motion field is extrapolated to the original reference frames R_{ k }. By construction, every pixel Y(p), p = (p_{1}, p_{2}) in the side information frame Y is located inside a number of overlapping blocks ${Y}_{{\mathbf{u}}_{n}}$ with u_{ n }= (u_{n,1}, u_{n,2}). After the execution of the OBME, a temporal predictor block ${R}_{k,{\mathbf{u}}_{n}}$ for every block ${Y}_{{\mathbf{u}}_{n}}$ has been identified in one reference frame. As a result, each pixel Y(p) in the side information frame has a number of associated temporal predictors ${r}_{k,{\mathbf{u}}_{n}}$ in the blocks ${R}_{k,{\mathbf{u}}_{n}}$.
However, some temporal predictors may stem from rather unreliable motion vectors. Especially when the input sequence was recorded at low frame rates or when the motion content is highly irregular, as might be the case in endoscopic sequences, temporal prediction is not the preferred method for all blocks at all times. Therefore, to avoid quality degradation of the side information due to untrustworthy predictors, all obtained motion vectors are subjected to a reliability screening. Namely, when the SAD, based on which the motion vector associated with temporal predictor ${r}_{k,{\mathbf{u}}_{n}}$ was determined, is not smaller than a certain threshold T, the motion vector and associated temporal predictor is labeled as unreliable. In this case, a temporal predictor for the side information pixel Y(p) is replaced by the colocated pixel of Y(p) in the upsampled hash frame, that is $\stackrel{\u0303}{W}\left(\mathbf{p}\right)$. In other words, when motion estimation is considered not to be trusted, the hash itself is assumed to convey more dependable information. This feature of OBME is referred to as hashpredictorselection (HPS).
where, ${N}_{k,{\mathbf{u}}_{c}}$ denotes the number of predictors for pixel Y(p) and ${g}_{k,{\mathbf{u}}_{n}}={r}_{k,{\mathbf{u}}_{n}}$ when ${r}_{k,{\mathbf{u}}_{n}}$ is reliable or ${g}_{k,{\mathbf{u}}_{n}}=\stackrel{\u0303}{W}\left(\mathbf{p}\right)$ when ${r}_{k,{\mathbf{u}}_{n}}$ is unreliable. The derived multihypothesis motion field is employed in an analogous manner to estimate the chroma components of the side information frame from the chroma components of the reference frames R_{ k } or the upsampled hash.
4.2.3. WynerZiv decoding
Thereafter, the estimated correlation channel statistics per coded DCT band bitplane are interpreted into soft estimates, i.e., loglikelihood ratios (LLRs). These LLRs, which provide a priori information about the probability of each bit to be 0 or 1, are passed to the variable nodes of the LDPCA decoder. Then, the message passing algorithm [41] is used for iterative LDPC decoding, in which the received syndrome bits correspond to the check nodes on the bipartite graph.
where equality in (6) stems from: p(b_{ l }y, b_{1}, ..., b_{l1}) = p(b_{1}, ..., b_{l1}, b_{ l }y)/p(b_{1}, ..., b_{l1}y). Hence, in (6) the nominator and the denominator are calculated by integrating the conditional probability density function of the correlation channel, i.e., f_{XY}(xy), over the quantization bin indexed by b_{1}, ..., b_{ l }.
Remark that the LDPCA decoder achieves various rates by altering the decoding graph upon reception of an additional increment of the accumulated syndrome [36]. Initially, the decoder receives a short syndrome based on an aggressive code and the decoder tries to decode [36]. If decoding falls short, the encoder receives a request to augment the previously received syndrome with extra bits. The process loops until the syndrome is sufficient for successful decoding.
where, q_{L}, q_{H} denote the lower and upper bound of the quantization bin q. Finally, the inverse DCT transform provides the reconstructed frame Ŵ in the spatial domain. The reconstructed frame is now ready for display and is stored in the reference frame buffer, serving as a reference for future temporal prediction.
5. Evaluation
The experimental results have been divided into three distinct parts. Namely, first the proposed system is compared against a set of relevant alternative video coding solutions using traditional test sequences. The second part comprises the experimental validation of our system in the application of wireless capsule endoscopy, comparing its performance against coding solutions currently used for the compression of endoscopic video. The third part elaborates on the encoding complexity of the proposed architecture.
We begin by defining the configuration elements of the proposed system, which are common to both types of input video. Namely, the motion estimation algorithm was configured with an overlap step size ε = 4, the size of the overlapping blocks was set to B = 16 and the threshold was chosen T = 400. The motion search was executed in an exhaustive manner at integerpel accuracy within a search range of ± 16 pixels. The downscaling factor to create the hash was fixed at d = 2.
5. 1. Evaluation on conventional test sequences
Employed quantization parameters for the key, the hash and the WynerZiv frames as well as the resulting RSD for the entire sequence
RD point 1 (QM1)  RD point 2 (QM4)  RD point 3 (QM7)  RD point 4 (QM8)  

Ice  
Key frame QP  40  34  29  25 
Hash QP  41  40  39  38 
RSD(%)  2.25  2.26  1.90  1.28 
Foreman  
Key frame QP  40  34  29  25 
Hash QP  41  40  39  38 
RSD(%)  2.92  2.97  2.58  1.96 
Silent  
Key frame QP  37  33  29  24 
Hash QP  40  39  38  37 
RSD(%)  2.29  1.02  0.54  2.38 
Soccer  
Key frame QP  44  36  31  25 
Hash QP  45  42  41  38 
RSD(%)  4.41  3.29  2.96  2.73 
To further evaluate the performance of our proposed scheme, the coding results of [12] are included in Figure 4. The hashbased WynerZiv video codec of [12] combines MCI with hashdriven motion estimation using low quality H.264/AVC Intra coded WynerZiv blocks to generate side information. Even though the codec of [12] advances over DISCOVER, our proposed hashbased solution generally exhibits higher performance bringing BD rate savings of 17.68 and 12.18% in Foreman and Soccer, in GOP8, respectively.
Lastly, the proposed DVC is compared with H.264/AVC Intra, which represents the lowcomplexity configuration of the stateoftheart traditional coding paradigm. One can observe from Figure 5 that in lowmotion sequences the proposed codec is superior to H.264/AVC Intra, bringing BD rate savings of up to 26.7% in Silent, GOP8. However, under difficult motion conditions like in Ice or Soccer H.264/AVC Intra is very efficient compared to DVC systems, which is in agreement with the results shown in Figures 4 and 5. We emphasize that the encoding complexity of H.264/AVC Intra is much higher than any of the presented DVC solutions, as discussed in Section 5.3.
5.2. Evaluation on endoscopic video sequences
A major contribution of this article is the assessment of WynerZiv coding for endoscopic video data, characterized by its unique content. In the proposed codec, the quantization parameters of the WynerZiv frames, the key frames, and the hash are meticulously selected so as to retain high and quasiconstant decoded frames' quality, as demanded by medical applications. Furthermore, in order to deliver highquality decoding under the strenuous conditions of highly irregular motion content and low frame acquisition rates, the proposed codec employs a GOP size of 2.
Initially, in order to prove the potential of its application in contemporary wireless capsule endoscopic technology, the proposed codec has been appraised using four capsule endoscopic test video sequences visualizing diverse areas of the gastrointestinal track. These sequences were extracted from extensive capsule endoscopic video material of two capsule examinations from two random volunteers^{f} performed at the Gastroenterology Clinic of the Universitair Ziekenhuis Brussels, Belgium. In the aforementioned clinical examinations, the capsule acquisition rate was two frames per second with a frame resolution of 256 × 256 pixels. The obtained test video sequences^{g} are termed "Capsule Test Video 1" to "Capsule Test Video 4" in the remainder of the article.
Figure 7 also evaluates the impact of the flexible scheme that enables the proposed OBME method to identify erroneous motion vectors and to replace the temporal predictor pixel with the decoded and interpolated hash. The results show that the proposed system with the HPS module remarkably advances over its equivalent that solely retains predictors from the reference frames. Specifically, in "Capsule Test Video 1" to "Capsule Test Video 4" adding the HPS functionality results in BD [42] rate improvements of 21.1, 16.02, 12.93, and 12.06%, respectively.
Future generations of capsule endoscopic technology aim at diminishing the quality difference with respect to conventional endoscopy by increasing the frame rate and resolution. Therefore, to confirm its capability under these conditions, the proposed WynerZiv video codec is evaluated using conventional endoscopic video sequences monitoring diverse parts of the digestive track of several patients. The endoscopic test video sequences considered in this experimental setting have a frame rate of 30 Hz and a frame resolution of 480 × 320 pixels. These endoscopic test video sequences are further referred to as "Endoscopic Test Video 1" to "Endoscopic Test Video 6". In this experiment, the proposed codec employs H.264/AVC Intra (Main profile) to code the key and the hash frames. Notice that the H.264/AVC Intra codec constitutes a recognized reference for medical video compression, e.g. [43].
Compared to H.264/AVC Intra, the experimental results in Figure 10 show that the proposed codec delivers BD rate savings of 4.1% in "Endoscopic Test Video 2". In "Endoscopic Test Video 1" and "Endoscopic Test Video 3" the proposed codec falls behind H.264/AVC Intra, incurring a BD rate loss of 3.84 and 0.20%, respectively. Only in "Endoscopic Test Video 4", which comprises highly irregular motion, the experienced Bjøntegaard rate overhead is notable amounting to 15.68%. Notice that the benefit of the HPS functionality of the proposed codec is reduced in case of conventional endoscopic video with respect to the capsule endoscopic sequences. This is due to the fact that the former sequences were recorded at a much higher frame rate and contain more temporal correlation. Nevertheless, in "Endoscopic Test Video 4" the HPS module brings BD rate savings of 4.63%.
5.3. Encoding complexity
Lowcost encoding is a key aspect of distributed video compression. During the evaluation of the DISCOVER [26] codec, it was shown that the WynerZiv frames' encoding complexity is very low compared to the complexity associated with the intra encoding of the key frames. Therefore, the lower the number of key frames, i.e., the longer the GOP, the higher the gain in complexity reduction offered by DVC over H.264/AVC Intra frame coding. Execution time measurements under controlled conditions, as established by the DISCOVER group [26], have shown that our codec (using H.264/AVC Intra to code the hash and the key frames) brings a reduction in average encoding time of approximately 30, 50, and 60% for a GOP size of 2, 4, and 8, respectively, compared to H.264/AVC Intra.
In contrast to hashless WynerZiv codecs, e.g. [26], our proposed codec has a higher encoding complexity caused by the additional hash formation and coding. However, the hashrelated complexity overhead is kept low, since the hash dimensions were reduced to one fourth of the original frame resolution prior to coarse H.264/AVC Intra frame coding. When compared to Motion JPEG, the proposed codec (although currently not optimized for speed) exhibits similar encoding time but offers superior compression performance. We remark that compared to DISCOVER or Motion JPEG, the proposed codec offers a significant reduction of the encoding rate for a given distortion level. Such a notable rate reduction induces an important decrease in power consumption by the transmission part of wireless video recording devices, e.g., wireless capsule endoscopes.
The proposed system links the encoder to the decoder via a feedback channel. Such a reverse channel implies that the encoder is forced to store WynerZiv data in a buffer pending the decoder's directives. Based on our prior work [44], we analyze of the buffer size requirements imposed on the presented system's encoder due to the decoding delay for the capsule endoscopy application scenario. Recall that the GOP size in this scenario is restricted to 2 frames (see Section 5.2). The prime factors determining the decoding delay are the frame acquisition period t_{ F }, the time t_{SI} to generate a side information frame, the transmission time (timeofflight) t_{TOF} between encoder and decoder, and the LDPC softinput softoutput decoding time, denoted by t_{SISO}.
Continuing our analysis, the reported capsule and the conventional endoscopic sequences were recorded using a camera with an acquisition rate of 2 and 30 Hz, respectively, corresponding to an acquisition period t_{ F } of 500 and 33.33 ms.
An estimation of the transmission time t_{TOF} through the body can be made by calculating the velocity of a uniform plane in a lossy medium [45], characterized by its dielectric properties, i.e. the conductivity and permittivity. These values can be calculated based on [46, 47] for a wide range of body tissues and frequencies. It can be verified that at a frequency of 433 MHz the velocity is always greater than 10% of the speed of light through all body tissue cases included in [47], leading to a timeofflight t_{TOF} in the order of 15 ns through 0.5 m of tissue.
It is clear that the time t_{SI} to generate a side information frame is dominated by OBME. Fortunately, several VLSI designs for hardware implementation of block motion estimation have been proposed. Considering the stateoftheart architecture of [48], full integerpel motion search can be executed at 4ρ^{2} + B1 cycles per macroblock (MB), where ρ and B are the search range and MB size, respectively. However, our presented scheme employs bidirectional OBME. Specifically, the total number of overlapping blocks per frame is (H·V)/ε^{2}, where H and V are the horizontal and vertical frame dimensions and ε is the overlap size. Hence, based on the VLSI architecture in [48], the total number of cycles per frame is given by 2·[(4·ρ^{2} + B1)·H·V]/ε^{2}, where the factor 2 stems from the bidirectionality. Considering a simplified decoding device with a single core CPU running at 800 MHz with a 1DIMPS/MHz/core^{h} and instantiating the OBME parameters for ρ = 16, ε = 4, and B = 16, yields a delay of 10.63 and 24.9 ms per frame for the capsule endoscopic (H = V = 256) and the endoscopic (H = 480, V = 320) sequences, respectively.
Average feedback channel requests per WynerZiv frame for the capsule endoscopic video sequences
RD point 1  RD point 2  RD point 3  RD point 4  

Capsule Test Video 1  35.4  41.2  59.3  86.6 
Capsule Test Video 2  33.7  39.7  59.6  85.4 
Capsule Test Video 3  34.1  39.4  55.8  80.9 
Capsule Test Video 4  32.7  40.5  59.0  88.3 
Based on the above approximations, Equation (8) yields an estimated buffer size of L = 2 and L = 3 frames for the capsule (t_{ F } = 500 ms) and the conventional endoscopic (t_{ F } = 33.33 ms) sequences, respectively, thus confirming the applicability of the proposed scheme. However, the encoder buffer size can be further restrained. An elegant solution is to constrain the number of feedback requests to a fixed number of requests for an entire WynerZiv frame as proposed in our previous study [44], where we show that the loss of compression efficiency compared to unconstrained feedback is less than 5% when at most F = 5 requests per WynerZiv frame are allowed. In addition to this, the structural latency induced by bidirectional temporal prediction could be reduced by employing unidirectional prediction.
6. Conclusions
Motivated by the strict prerequisites of wireless lightweight multimedia applications, such as wireless capsule endoscopy, this article has introduced a novel video codec based on the principles of WynerZiv coding. The proposed codec maintains low encoding complexity and facilitates quality and temporal scalability. Intrinsically, the proposed codec achieves high compression performance by embracing a novel hashdriven motion estimation technique, which generates accurate side information at the decoder. The presented technique performs motioncompensated multihypothesis prediction, enabling adaptation to the regional variations in temporal correlation in a frame. Concrete experimentation using various conventional and endoscopic test video sequences has confirmed the superior compression performance of the proposed codec against several stateoftheart traditional and WynerZiv video codecs. In effect, in conventional and endoscopic test video material significant Bjøntegaard rate savings of up to 32.13 and 43.37% over the stateoftheart have been obtained.
Endnotes
^{a}This paper has been presented in part at the IEEE International Conference on Image Processing, Brussels, Belgium, September 2011 [51]. ^{b}Motion JPEG is based on the JPEG coding standard [52] and includes a file format that can handle multiple JPEG images. Unlike the Motion JPEG 2000 standard [53], no standard specification has been defined for Motion JPEG, and hence only proprietary solutions are available (e.g., support in Microsoft AVI files, Apple Quicktime format or the RFC 2435 spec that describes how Motion JPEG can be supported by an RTP stream). ^{c}To simplify the presentation, the LDPC code is assumed systematic. ^{d}The experimental results of DISCOVER [26] have been obtained using the executable of the DISCOVER codec which is available on the projects website [26]. ^{e}Given an RD point and a number of iterations, the process starts from a specific hash QP value (QP_hash = QP_key+1), and calculates the total and the hash rate, and the resulting RSD of the decoded frames (both key and WZ frames). If the RSD is lower that a strict threshold, the QP and the rate values are stored; otherwise they are discarded. Next, the hash QP is increased and the algorithm continuous till it reaches a given number of iterations. Out of the retained QPs, the one which minimizes the total rate is chosen as the best for the specific rate point. In case of equal total rates the highest QP value is selected. ^{f}These volunteers presented no evidence of gastrointestinal pathologies. ^{g}These sequences were transformed to the YUV 4:2:0 format supported by the proposed codec. ^{h}Nowadays more powerful processors exist to be deployed in devices at the size of the decoder of a capsule endoscope. For instance, the Apple A5 processor of iPhone 4S has an ARM CortexA9 MPCore 32bit multicore processor at 800 MHz, 2.5DIMPS/MHz/core. ^{i}All these figures correspond to SoftInput Soft Output (SISO) decoders.
Abbreviations
 AVC:

advanced video coding
 CPU:

central processing unit
 CRC:

cyclic redundancy check
 DCT:

discrete cosine transform
 DISCOVER:

distributed coding for video services
 DJSCC:

distributed joint sourcechannel coding
 DVC:

distributed video coding
 DSC:

distributed source coding
 GOP:

group of pictures
 HPS:

hashpredictorselection
 JPEG:

joint photographic experts group
 LDPCA:

lowdensity paritycheck accumulate
 LLR:

log likelihood ratio
 MCI:

motioncompensated interpolation
 OBME:

overlapped block motion estimation
 PRISM:

powerefficient robust highcompression syndromebased multimedia coding
 PSNR:

peak signaltonoise ratio
 RSD:

relative standard deviation
 QCIF:

quarter common intermediate format
 QM:

quantization matrix
 QP:

quality parameter
 RD:

ratedistortion
 SAD:

sum of the absolute differences
 TDWZ:

transformdomain WynerZiv.
Declarations
Acknowledgements
This study was supported by the FWO Flanders projects G.0391.07, G.0146.10 and the postdoctoral fellowship of Peter Schelkens. The authors would like to thank Prof. Dr. Daniel Urbain, head of the Gastroenterology clinic of the Universitair Ziekenhuis Brussel for providing numerous endoscopic video sequences.
Authors’ Affiliations
References
 Wiegand T, Sullivan GJ, Bjøntegaard G, Luthra A: Overview of the H.264/AVC video coding standard. IEEE Trans Circ Syst Video Technol 2003, 13(7):560576.View ArticleGoogle Scholar
 Slepian D, Wolf JK: Noiseless coding of correlated information sources. IEEE Trans Inf Theory 1973, 19(4):471480. 10.1109/TIT.1973.1055037MathSciNetView ArticleMATHGoogle Scholar
 Wyner AD, Ziv J: The ratedistortion function for source coding with side information at the decoder. IEEE Trans Inf Theory 1976, 22(1):110. 10.1109/TIT.1976.1055508MathSciNetView ArticleMATHGoogle Scholar
 Puri R, Ramchandran K: PRISM: a new robust video coding architecture based on distributed compression principles. In 40th Allerton Conference on Communication, Control, and Computing. Volume 40. Allerton, IL; 2002:586595.Google Scholar
 Girod B, Aaron A, Rane S, RebolloMonedero D: Distributed video coding. Proc IEEE 2005, 93(1):7183.View ArticleMATHGoogle Scholar
 Esmaili GR, Cosman PC: WynerZiv video coding with classified correlation noise estimation and key frame coding mode selection. IEEE Trans Image Process 2011, 20(9):24632474.MathSciNetView ArticleGoogle Scholar
 Deligiannis N, Barbarien J, Jacobs M, Munteanu A, Skodras A, Schelkens P: Sideinformationdependent correlation channel estimation in hashbased distributed video coding. IEEE Trans Image Process 2012, in press.Google Scholar
 TotoZarasoa V, Roumy A, Guillemot C: Nonuniform source modeling for distributed video coding. In European Signal Processing Conference. Aalborg, Denmark; 2010:18891893.Google Scholar
 Pereira F, Torres L, Guillemot C, Ebrahimi T, Leonardi R, Klomp S: Distributed video coding: selecting the most promising application scenarios. Signal Process Image Commun 2008, 23(5):339352. 10.1016/j.image.2008.04.002View ArticleGoogle Scholar
 Stankovic L, Stankovic V, Cheng S: Distributed compression: overview of current and emerging multimedia applications. In IEEE International Conference on Image Processing. Brussels, Belgium; 2011:18411844.Google Scholar
 Sidhu R, Sanders D, McAlindon M: Gastrointestinal capsule endoscopy: from tertiary centres to primary care. BMJ 2006, 332: 528531. 10.1136/bmj.332.7540.528View ArticleGoogle Scholar
 Ascenso J, Brites C, Pereira F: A flexible side information generation framework for distributed video coding. Multimedia Tools Appl 2010, 48(3):381409. 10.1007/s1104200903166View ArticleGoogle Scholar
 Deligiannis N, Jacobs M, Barbarien J, Verbist F, Škorupa J, Van de Walle R, Skodras A, Schelkens P, Munteanu A: Joint DC coefficient band decoding and motion estimation in WynerZiv video coding. In International Conference on Digital Signal Processing. Corfu, Greece; 2011:16.Google Scholar
 Shannon CE: Coding theorems for a discrete source with a fidelity criterion. IRE National Convention Record 1959, 142163.Google Scholar
 Wyner A: Recent results in the Shannon theory. IEEE Trans Inf Theory 1974, 20(1):210. 10.1109/TIT.1974.1055171MathSciNetView ArticleMATHGoogle Scholar
 Pradhan S, Ramchandran K: Distributed source coding using syndromes (DISCUS): design and construction. IEEE Trans Inf Theory 2003, 49: 626643. 10.1109/TIT.2002.808103MathSciNetView ArticleMATHGoogle Scholar
 Mitran P, Bajscy J: Coding for the WynerZiv problem with turbolike codes. In IEEE International Symposium on Information Theory. Lausanne, Switzerland; 2002:91.View ArticleGoogle Scholar
 Liveris A, Xiong Z, Georghiades C: Compression of binary sources with side information at the decoder using LDPC codes. IEEE Commun Lett 2002, 6(10):440442.View ArticleGoogle Scholar
 Zamir R, Shamai S, Erez U: Nested linear/lattice codes for structured multiterminal binning. IEEE Trans Inf Theory 2002, 48(6):12501276. 10.1109/TIT.2002.1003821MathSciNetView ArticleMATHGoogle Scholar
 Liu Z, Cheng S, Liveris A, Xiong Z: SlepianWolf coded nested quantization (SWCNQ) for WynerZiv coding: performance analysis and code design. In IEEE Data Compression Conference. Snowbird, UT; 2004:322331.Google Scholar
 Chou J, Pradhan S, Ramchandran K: Turbo and trellisbased costructions for source coding with side information. In IEEE Data Compression Conference. Snowbird, UT; 2003:3342.Google Scholar
 Yang Y, Cheng S, Xiong Z, Zhao W: WynerZiv coding based on TCQ and LDPC codes. IEEE Trans Commun 2009, 57(2):376387.View ArticleGoogle Scholar
 Xu Q, Stankovic V, Xiong Z: Layered WynerZiv video coding for transmission over unreliable channels. Signal Process Elsevier 2006, 86: 32123225.View ArticleMATHGoogle Scholar
 Aaron A, Rane S, RebolloMonedero D, Girod B: Systematic lossy forward error protection for video waveforms. In International Conference on Image Processing. Volume I. Barcelona, Spain; 2003:609612.Google Scholar
 Cheng S, Xiong Z: Successive refinement for the WynerZiv problem and layered code design. IEEE Trans Signal Process 2005, 53(8):32693281.MathSciNetView ArticleGoogle Scholar
 Artigas X, Ascenso J, Dalai M, Klomp S, Kubasov D, Quaret M: The DISCOVER codec: architecture, techniques and evaluation. In Picture Coding Symposium. Lisboa, Portugal; 2007. [http://www.discoverdvc.org] []Google Scholar
 Ascenso J, Brites C, Pereira F: Content adaptive WynerZiv video coding driven by motion activity. In IEEE International Conference on Image Processing. Atlanta, GA; 2006:605608.Google Scholar
 Aaron A, Rane S, Girod B: WynerZiv video coding with hashbased motion compensation at the receiver. In IEEE International Conference on Image Processing. Volume 5. Singapore; 2004:30973100.Google Scholar
 Dinh TN, Lee G, Chang JY, Cho HJ: Side information generation using extra information in distributed video coding. In IEEE International Symposium on Signal Processing and Information Technology. Cairo, Egypt; 2007:138143.Google Scholar
 Deligiannis N, Munteanu A, Clerckx T, Cornelis J, Schelkens P: Overlapped block motion estimation and probabilistic compensation with application in distributed video coding. IEEE Signal Process Lett 2009, 16(9):743746.View ArticleGoogle Scholar
 Verbist F, Deligiannis N, Jacobs M, Barbarien J, Schelkens P, Munteanu A: A statistical approach to create side information in distributed video coding. ACM/IEEE International Conference on Distributed Smart Cameras 2011.Google Scholar
 Deligiannis N, Jacobs M, Verbist F, Slowack J, Barbarien J, Van de Walle R, Schelkens P: A Munteanu, Efficient hashdriven WynerZiv video coding for visual sensors. In ACM/IEEE International Conference on Distributed Smart Cameras. Ghent, Belgium; 2011.Google Scholar
 McCaffrey C, Chevalerias O, O'Mathuna C, Twomey K: Swallowablecapsule technology. IEEE Pervasive Comput 2008, 7(1):2329.View ArticleGoogle Scholar
 Turgis D, Puers R: Image compression in video radio transmission for capsule endoscopy. Sensors Actuat A: Physical 2005, 123124: 129136.View ArticleGoogle Scholar
 Duchon CE: Lanczos filtering in one and two dimensions. J Appl Meteorol 1979, 18(8):10161022. 10.1175/15200450(1979)018<1016:LFIOAT>2.0.CO;2View ArticleGoogle Scholar
 Varodayan D, Aaron A, Girod B: Rateadaptive codes for distributed source coding. Signal Process Elsevier 2006, 86: 31233130.View ArticleMATHGoogle Scholar
 Abouei J, Brown JD, Plataniotis KN, Pasupathy S: Energy efficiency and reliability in wireless biomedical implant systems. IEEE Trans Inf Technol Biomed 2011, 15(3):456466.MathSciNetView ArticleGoogle Scholar
 Bradley PD: An ultra low power, high performance Medical Implant Communication System (MICS) transceiver for implantable devices. In IEEE Biomedical Circuits and Systems Conference. London, UK; 2008:158161.Google Scholar
 Oppenheim AV, Schafer RW, Buck JR: DiscreteTime Signal Processing. 2nd edition. Prentice Hall, Upper Saddle River; 1999.Google Scholar
 Turkowski K, Gabriel S: Filters for common resampling tasks. In Graphics Gems I. Edited by: Glassner AS. Academic Press, San Diego; 1990.Google Scholar
 Chung S, Richardson T, Urbanke R: Analysis of sumproduct decoding of lowdensity paritycheck codes using a Gaussian approximation. IEEE Trans Inf Theory 2001, 47(2):657670. 10.1109/18.910580MathSciNetView ArticleMATHGoogle Scholar
 Bjøntegaard G: Calculation of average PSNR differences between RDcurves. 2001.Google Scholar
 Yu H, Lin Z, Pan F: Applications and improvement of H.264 in medical video compression. IEEE Trans Circ Syst I 2005, 52(12):27072716.View ArticleGoogle Scholar
 Slowack J, Skorupa J, Deligiannis N, Lambert P, Munteanu A, Van de Walle R: Distributed video coding with feedback channel constraints. IEEE Trans Circ Syst Video Technol 2012, in press.Google Scholar
 Balanis CA: Advanced Engineering Electromagnetics. John Wiley and Sons, Inc., New York; 1989.Google Scholar
 Gabriel C: Compilation of the dielectric properties of body tissues at RF and microwave frequencies. Occupational and environmental health directorate, Radiofrequency Radiation Division, Brooks Air Force Base, TX; 1996.Google Scholar
 [http://niremf.ifac.cnr.it/tissprop/htmlclie/htmlclie.htm]
 Chen CY, Chien SY, Huang YW, Chen TC, Wang TC, Chen LG: Analysis and architecture design of variable blocksize motion estimation for H.264/AVC. IEEE Trans Circ Syst I 2006, 53(2):578592.View ArticleGoogle Scholar
 Brack T, Alles M, LehnigkEmden T, Kienle F, When N, L'Insalata NE, Rossi F, Rovini M, Fanucci L: Low complexity LDPC code decoders for next generation standards. In Design, Automation, and Test in Europe. Nice, France; 2007:16.Google Scholar
 Sun Y, Cavallaro JR, Ly T: Scalable and low power LDPC decoder design using high level algorithmic synthesis. In IEEE International SystemonChip Conference. Belfast, Northern Ireland, UK; 2009:267270.Google Scholar
 Deligiannis N, Verbist F, Barbarien J, Slowack J, Van de Walle R, Schelkens P, Munteanu A: Distributed coding of endoscopic video. In IEEE International Conference on Image Processing. Brussels, Belgium; 2011:18531856.Google Scholar
 Pennebaker WB, Mitchell JL: JPEG Still Image Data Compression Standard. Van Nostrand Reinhold, New York; 1993.Google Scholar
 Schelkens P, Skodras A, Ebrahimi T: The JPEG 2000 Suite. Wiley, Chichester; 2009.View ArticleGoogle Scholar
Copyright
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.