H. Sun and W. Kwok, "Concealment of damaged block transform coded image using projections onto convex sets," IEEE Trans. Image Processing, vol. 4, no. 4, pp. 470--477, Apr. 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... waveform coding, i.e. redundancy in the waveform coding stage [6] 5) Robust entropy coding, which places redundancy in the entropy coding stage [11, 13] 6) Postprocessing at the decoder, which blindly recover the lost pixels based on their surrounding spatial or temporal neighborhood pixels [3, 15]. In [20] Yin et al. show that the quality of blind recovery can be enhanced by embedding additional features in the image. For instance, they embed the edge directionality information bit of blocks into images. In this paper, we propose a new method of using our ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Trans. on Image Processing, Vol. 4, pp. 470-477, Apr. 1995.

.... concealment may fail in the presence of scene changes, fast motion, rotation and deformation of objects [2] Therefore, a practical algorithm should also include a spatial interpolator to recover missing macroblocks (MB) e.g. based on a DCT domain formulation [3] or on projection onto convex sets [4]. Spatio temporal techniques have also been proposed, using boundary matching [2] or Markov random fields (MRF) 5, 6, 7] In most spatio temporal error concealment algorithms, once a corrupted MB or slice has been detected, one has to decide whether to apply spatial or temporal restoration. For ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Transactions on Image Processing, vol. 4, no. 4, pp. 470--477, Apr. 1995. (c) (b) (d) Fig. 6. An inter-coded frame from the foreman sequence: (a) original; reconstruction using (b) F-MS, (c) MB-MS, (d) GBM-MS

....spatial error concealment is an invaluable resource. As for spatial error concealment, several techniques have been proposed. The problem has been formulated in the DCT domain [2] where smoothness of the recovered MB can be easily imposed. Projection onto convex sets has also been suggested [3]. More recent approaches consider edge preserving interpolators. Amongst others, in [4] it is proposed to estimate edge directions by means of the Hough transform while in [5] an approach based on sketch theory is taken on, consisting in the recovery of contours followed by bilinear interpolation ....

H. Sun, W. Kwok, "Concealment of damaged block transform coded images using projection onto convex sets", IEEE Transactions on Image Processing, vol. 4, no. 4, pp. 470-477, Apr. 1995.

.... domain, but also to the spectral domain such as the discrete cosinetransform (DCT)domain, as proposedby Chung et al. 16] Some other methods are based on projection onto convex sets (POCS) that iteratively uses the smoothnessassumptionand pixel orDCT value range information forerrorconcealment [17][18] Anextensiontotheassumption thatnaturalimages are smooth and the values are continuous spatially or spectrally is to adopt the Markov random fields (MRF)to model the images [19] MRF based error concealment methods were first proposedby Salama et al. 20] 22] LaterShirani ....

H. Sub and W. Kwok, Concealment of Damaged Block Transform Coded Images using Projections OntoConvex Sets , IEEETrans.onImage Processing,Vol.4,pp.470- 477,April1995.

....but not diagonal edges. Another assumption made about images is edge continuity, which implies that if an edge is present in the neighboring blocks of a missing block, and its direction indicates that it should passes through the missing block, it should be then extended to the missing block. In [7], this implication is used in a Projection Onto Convex Sets (POCS) based retrieval method. In [8] a heuristic approach which is less computationally intensive, replaces the POCS stage in the above method. Although generally more accurate than the approach proposed in [3] edge based methods are ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using pro- jection onto convex sets," IEEE Transactions image processing, vol. 4, pp. 470 477, April 1995.

....in computation by suitable bu#ering and updating of transform coe#cients. Note also that if one views the iterative use of denoising and projection onto layers as a sequence of alternating projections, it can be seen that the proposed method is substantially di#erent from approaches such as [10], since the operation of the presented algorithm cannot be simplified to a sequence of projections onto fixed, data independent, convex or nonconvex sets. Further issues related to the sensitivity to compression, computational complexity, selection of thresholds, and other results on the optimum ....

H. Sun, and W. Kwok, "Concealment of Damaged Block Transform Coded Images Using Projections onto Convex Sets," IEEE Trans. on Image Proc., Vol. 4, No. 4, Aug. 1995.

....GOB, or at least a significant part of it, is affected (i.e. 16 successive luminance lines) spatial interpolation is less efficient than temporal extrapolation. Only in the case of very complex motion or scene cuts, it can be advantageous to rely on the spatial correlation in the image [37] [38], or switch between temporal and spatial concealment [39] 32] In the simplest and most common approach, previous frame concealment, the corrupted image content is replaced by corresponding pixels from the previous frame. This simple approach yields good results for sequences with little motion ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projection onto convex sets," IEEE Trans. Image Processing, vol. 4, pp. 470--477, Apr. 1995.

....communication, intra frame coding (thus spatial interpolation for error concealment) may be preferred in order to avoid severe inter frame error propagation and to reduce the hardware complexity. Various methods have been proposed to conceal the missing data by exploiting the spatial redundancy [3, 4, 5, 6]. Most of these schemes are related to block based image and video coding techniques, probably because the most emerging image and video compression standards such as JPEG [7] H.261 [8] and MPEG [9] are all block based coding schemes. Wang et al. 3] imposed spatial and temporal smoothing ....

....al. 3] imposed spatial and temporal smoothing constraints on the reconstructed block to estimate the lost DCT coefficients. Block interleaving was incorporated to minimize the effect of packet loss so that the damaged blocks tend to be isolated and the loss recovery is facilitated. Sun and Kwok [5] utilized spatially correlated edge information from a large local neighborhood of surrounding pixels to conceal damaged blocks. Their approach can be viewed as an alternating projection onto convex set method. A fuzzy logic reasoning based approach [6] was proposed by Lee et al. for block based ....

[Article contains additional citation context not shown here]

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Trans. Image Proc., pp. 470-477, April 1995.

....from transmission errors is minimized. Error concealment is an extremely important component of any error robust video codec. Spatial and temporal interpolations are often utilized in error concealment methods. Examples include Maximally Smooth Recovery ( 11] Projection onto Convex Sets, [12]) and various motion vector and coding mode recovery methods such as Motion Compensated Temporal Prediction ( 13] Like the error resilience tools discussed above, the effectiveness of an error concealment strategy is highly dependent on the performance of the resynchronization scheme. If the ....

H. Sun, W. Kwok, "Concealment of damaged block transform coded images using projection onto convex sets," IEEE Trans. Image Processing, vol. 4, pp. 470-477, Apr. 1995.

....HDTV applications [6] and Jung et al. proposed an EC algorithm in the spatial domain by using the projective interpolation [7] 84 Y. L. Huang and R. F. Chang Also, Sun and Kwok proposed a spatial interpolation algorithm for the EC problem based on the projections onto convex set (POCS) theory [8]. We notice that the spatial projective interpolation schemes utilise different reconstruction rules that are decided by the edge pattern. If the block size is large or the edges in a block are irregular, the edge classification scheme will become complex and inefficient, and then the projective ....

Sun H, Kwok W. Concealment of damaged block transform coded images using projections on convex sets. IEEE Trans Image Process 1995; 4: 470--477

....while temporal interpolation is used to reconstruct the missing data in inter coded frames. In recent years, numerous error concealment schemes have been proposed in the literature (refer to Ref. 69] for a good survey) Examples include maximally smooth recovery [67] projection onto convex sets [58], and various motion vector and coding mode recovery methods suchas motion compensated temporal prediction [25] So far, we have reviewed various application layer QoS control techniques. These techniques are employed by the end systems and do not require any QoS support from the network. If the ....

H. Sun and W. Kwok, \Concealment of damaged block transform coded images using projection onto convex sets," IEEE Trans. on Image Processing, vol. 4, no. 4, pp. 470-477, April 1995.

....and the solutions under different loss patterns correspond to different interpolation filters in the spatial, temporal, and frequency domains. Spatial Interpolation Using Projection onto Convex Sets (POCS) Technique: Another way of accomplishing spatial interpolation is by using the POCS method [ 38 , 39 ]. The general idea behind POCSbased estimation methods is to formulate each constraint about the unknowns as a convex set. The optimal solution is the intersection of all the convex sets, which can be obtained by recursively projecting a previous solution onto individual convex sets. When applying ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Trans. Image Proc., vol. 4, no. 4, pp. 470-477, Apr. 1995.

....for real time applications since they require retransmission of ATM cells. Prioritization approaches to ATM cell loss concealment have been proposed [4, 5, 6] Techniques involving interleaving data have also been proposed [7, 8] along with post processing techniques for error concealment [9, 10, 7, 11, 12, 13]. In all of the above techniques there has been no mention of how the loss of macroblocks is detected. Furthermore, there been no reference as to how bits pertinent to one macroblock are distinguished from those belonging to another macroblock in the event that cell loss has occurred. Our approach ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Transactions on Image Processing, vol. 4, no. 4, pp. 470--477, April 1995.

....its neighboring slices kept intact. In [47] each frame is split into four bands and the data for each band packed together. This distributes the error due to cell loss over the entire frame, rather than localize it. Post processing techniques for the sake of error concealment are addressed in [48, 49, 46, 50, 51, 52, 53, 54, 55, 56, 57]. In [48, 49, 46] it is assumed that partial loss of DCT coe#cients has occurred. High frequency coe#cients that have been lost are replaced by zeros. The remaining coe#cients are estimated, by minimizing a cost function. The cost function is the sum, over all pixels in a block, of the square of ....

....method is based on a measure of image activity. If there is no predominant motion within a frame then temporal replacement is used, otherwise the macroblock is reconstructed by means of spatial interpolation. The method of projections onto convex sets [58] for error concealment is described in [53]. To reconstruct a missing macroblock, it is first determined whether an edge actually a passes through it. This is achieved by applying the Sobel operator [23] to the surrounding macroblocks. A larger block of pixels is then constructed from initial pixel values of the lost macroblock and those ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Transactions on Image Processing, vol. 4, no. 4, pp. 470--477, April 1995.

....the interpolation does not produce satisfactory results. Other approaches employ some combinations of the above three techniques to reconstruct lost data. Maximumsmoothness recovery extends the smoothness property to both spatial and temporal neighbors [85] POCS (Projection onto Convex Sets) [60, 81] formulates spatial and temporal smoothness constraints into convex sets and derives a solution iteratively. The approach using genetic algorithm conceals a corrupted block by iteratively performing reproduction crossover mutation and by evaluating a proposed fitness function until a stopping ....

H. Sun and W. Kwok. Concealment of damaged block transform coded images using projections onto convex sets. IEEE Trans. on Image Processing, 4(4):470-477, Apr. 1995.

....ave Dsd1c p ( Prob s p ( c p s ( Prob s 1 ( 8 JPEG are briefly described. The RVLC JPEG solution is then presented, and error detection and correction for the RVLC JPEG stream are outlined. Much work has been performed on error detection and or concealment for JPEG images [15 17]. This section presents a computationally simple, high quality detection and concealment strategy allowing evaluation of the performance of RVLC JPEG. A. JPEG and VLC Decoding Errors JPEG image compression is a 3 step operation consisting of applying a discrete cosine transform to pixel blocks, ....

H. Sun, W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE. Trans Image Proc., Vol. 4, April 1995, pp. 470-77.

.... or the smoothing of source coding artifacts, but the recovery of lost data does not seem so much covered and is in most of the cases related to block transforms (e.g. JPEG based systems) When some of the blocks are lost, the authors try to recover them either by global iterative algorithms [1, 2] or by various computations at the local level, aimed at producing a kind of interpolation from the neighbours [3,4,5] In our case, we are using a 3 octave wavelet transform [6] The transmitted images are thus coded into one low pass subband (DC subband) giving a down sampled approximation of ....

Huifang Sun and Wilson Kwok. Concealment of damaged block transform coded images using projections onto convex sets. IEEE Transactions on Image Processing, 4(5):470--477, April 1995.

....strategy under the assumptions which we have made. Previous work on concealment have exploited the prior information which we are using in this paper. Wang, Zhu and Shaw [5] assumed that the image is smooth and find a reconstruction which minimizes sudden changes in the image. Sun and Kwok [4] assumed that the image is either smooth or has a strong edge and proposes a method based on projection on convex sets for reconstruction of the missing blocks. For concealment in video, Zhang, Arnold and Frater [6] assumed that the current block is a translation of a block from the previous frame ....

H. Sun and W. Kwok. Concealment of damaged block transform coded images using projections onto convex sets. IEEE Trans. Image Processing, 4(4):470--477, April 1995.

....edges or lines are attempted to be reconstructed. In all other cases its performance deteriorates followed by noticeable blurring in the concealed regions. More complicated interpolation methods or spatial techniques have also been proposed in the literature [9] 14] 16] 26] 30] 32] [34] [39] 41] 42] to deal with these drawbacks. They are, however, usually more computationally intensive than the bilinear interpolation approach. Novel methods for the concealment of lost MBs in the first I frame of an MPEG 2 compressed sequence are investigated in this paper. They attempt to ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets", IEEE Trans. on Image Processing, vol. 4, no. 4, pp. 470--477, 1995.

....be studied. In particular it is important to study the e#ect of ATM cell loss, and develop post processing techniques that can be used for error concealment. There are two approaches to post processing the received data for the purpose of error concealment namely spatial and temporal interpolation [1 7]. In spatial interpolation, missing pixel values are reconstructed from neighboring spatial information, whereas in temporal interpolation the lost data is reconstructed from past and future frames via motion vectors that have to be estimated. This work was partially supported by a grant from the ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Transactions on Image Processing, vol. 4, no. 4, pp. 470--477, April 1995.

....GOB, or at least a significant part of it, is affected (i.e. 16 successive luminance lines) spatial interpolation is less efficient than temporal extrapolation. Only in the case of very complex motion or scene cuts, it can be advantageous to rely on the spatial correlation in the image [37] [38], or switch between temporal and spatial concealment [39] 32] In the simplest and most common approach, previous frame concealment, the corrupted image content is replaced by corresponding pixels from the previous frame. This simple approach yields good results for sequences with little motion ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projection onto convex sets," IEEE Trans. Image Processing, vol. 4, pp. 470-477, Apr. 1995.

....at the slice level, and in [18] each frame is split into four bands and the data for each band packed together. Interleaving distributes the error due to cell loss over the entire frame, rather than localize it. Post processing techniques for the sake of error concealment are addressed in [19, 20, 17, 21, 22, 23, 24]. In [19, 20, 17, 21] reconstruction of a missing macroblock is performed by estimating the low frequency DCT coe#cients from the DCT coe#cients of the neighboring macroblocks and ignoring the high frequency coe#cients. An approach that estimates missing edges in each block from edges in the ....

....blocks was proposed in [23] For each direction of an estimated edge, a version of the lost block is reconstructed by performing a number of one dimensional interpolations in the direction parallel to each edge. A final version of the missing block is obtained by merging all versions. In [24], the method of projections onto convex sets [25] was used to reconstruct missing macroblocks. In [22, 26, 27, 28, 29] lost macroblocks are constructed either by temporal replacement or spatial interpolation. The decision to use a particular method is based on a measure of image activity. If ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Transactions on Image Processing, vol. 4, no. 4, pp. 470--477, April 1995.

....error Recovery error Figure 6: The goal of error concealment is to recover reduce the lost using the available information. Although some damages are remained after error concealment, the overall damages are less than those damages without error concealment (cf. Figure 3) tial interpolation [9, 10]. There are also some temporal error concealment techniques. One is to simply replace the lost region from the corresponding regions of the previous frame. Another is to replace the lost region from the affine transformation of the previous frame [11] Still another is to have temporal error ....

H. Sun and W. Kwok, "Concealment of Damaged Block Transform Coded Images Using Projections onto Convex Sets," IEEE Trans. on Image Processing, vol. 4, no. 4, pp. 470--477, Apr. 1995.

....that in many cases, a portion of an image can be faithfully reconstructed from its neighboring region by utilizing the image spatial contiguity property. The image spatial contiguity property has been extensively used for error concealment in block based image and video coding and transmission [7, 8]. In [8] it was shown that with a nonlinear directional filter applied to a lost image block, the received damaged image can be reconstructed faithfully with block lost rates of as high as 50 . This property has been used to develop an economic way to transmit coded images for low bit rate coding ....

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Trans. Image Proc., pp. 470-477, April 1995.

....#################### #################### #################### Region of acceptable images Convex set 1 Convex set 2 Noisy image . Projections Figure 2: Pictorial representation of POCS. See section 3 for details. band limits must be known. In addition, recent extensions like [10] require expensive calculations of lines that intersect noise pixels. In other recent work [1] shows how to reduce band limited interpolation and extrapolation problems for finite dimensional signals to solution of a set of linear equations. Further, they show that the corresponding matrix is ....

SUN, H. and KWOK, W. "Concealment of Damaged Block Transform Coded Images Using Projections onto Convex Sets", IEEE Trans. Image Proc. , 4, No. 4, April 1995.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded image using projections onto convex sets," IEEE Trans. Image Processing, vol. 4, no. 4, pp. 470--477, Apr. 1995.

No context found.

H. Sun and W. Kwok. Concealment of Damaged Block Transform Coded Images Using Projections onto Convex Sets. IEEE Transactions on Image Processing, 4(4):470--477, April 1995.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded image using projections onto convex sets," IEEE Trans. Image Processing, vol. 4, no. 4, pp. 470--477, Apr. 1995.

No context found.

H. Sun and W. Kwok. Concealment of Damaged Block Transform Coded Images Using Projections onto Convex Sets. IEEE Transactions on Image Processing, 4(4):470--477, April 1995.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projection onto convex set," IEEE Trans. Image Processing, vol. 4, pp. 470--477, Apr. 1995.

No context found.

H. Sun, W. Kwok, Concealment of damaged block transform coded images using projection onto convex sets, IEEE Trans. Image Processing 4 (4) (April 1995) 470}477.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Trans. Image Proc., vol. 4, no. 4, pp. 470-477, Apr. 1995.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Trans. Image Processing, vol. 4, pp. 470--477, Apr. 1995.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Trans. Image Proc., vol. 4, no. 4, p. 470-77, April 1995.

No context found.

H. Sun and W. Kwok, "Concealment of Damaged Block Transform Coded Images Using Projections onto Convex Sets", IEEE Trans. Image Proc., vol. 4, no. 4, pp. 470-477, April 1995.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projection onto convex sets," IEEE Trans. Image Processing, vol. 4, pp. 470--477, Apr. 1995.

No context found.

H. Sun, W. Kwok, Concealment of damaged block transform coded images using projection onto convex sets, IEEE Trans. Image Processing 4 (4) (April 1995) 470}477.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Trans. Image Processing,vol. IP-4, pp. 470--477, Apr. 1995.

No context found.

H. Sun and W. Kwok, "Concealment of damaged block transform coded images using projections onto convex sets," IEEE Transactions on Image Processing, vol. 4, no. 4, pp. 470--477, April 1995.

No context found.

H. Sun and W. Kwok, "Concealment of Damaged Block Transform Coded Images Using Projections onto Convex Sets", IEEE Trans. on Image Processing, vol. 4, no. 4, pp. 470-477, April 1995.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC