Trường Nguyễn - Academia.edu (original) (raw)
Papers by Trường Nguyễn
2008 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video, 2008
Two new attacks on multiview videos in the view space is investigated, which is also applicable t... more Two new attacks on multiview videos in the view space is investigated, which is also applicable to stereo and free-view television. The first attack involves the generation of new views from different viewpoints. The second attack involves change in the region of focus (the display plane or the zero disparity plane) in the multiview images. A scale invariant feature descriptor (SIFT) based fingerprinting mechanism which can identify such attacks is developed. An online verification system matches the SIFT descriptors of the original video (stored at a central database) to that of the attacked video. Results show that the method can detect such attacks well, and is useful for copy detection on the internet.
2008 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video, 2008
A procedure for obtaining high dynamic range (HDR) videos from multiple differently exposed image... more A procedure for obtaining high dynamic range (HDR) videos from multiple differently exposed image sequences from a camera array is explored. It is observed that using information along both viewspace (camera) and temporal axes results in a good estimate of the HDR image sequence. Images captured at longer exposures are subject to motion blur. A novel motion deblurring scheme is proposed, prior to the actual HDR mapping process. This involves a multiscale directional structure preservation procedure which uses information from adjacent views along camera-space and frames along time. The proposed deblurring scheme works in spite of illumination variations between images.
2007 IEEE International Conference on Image Processing, 2007
The proposed algorithm in this work provides superresolution for color images by using a learning... more The proposed algorithm in this work provides superresolution for color images by using a learning based technique that utilizes both generative and discriminant approaches. The combination of the two approaches is designed with a stochastic classification-regression frame- work where a color image patch is first classified by its content, and then, based on the class of the patch, a learned
2008 IEEE International Conference on Acoustics, Speech and Signal Processing, 2008
In this paper, we apply the primal-dual decomposition and subgradi- ent projection methods to sol... more In this paper, we apply the primal-dual decomposition and subgradi- ent projection methods to solve the rate-distortion optimization prob- lem with the constant bit rate constraint. The primal decomposition method enables spatial or temporal prediction dependency within a Group Of Picture (GOP) to be processed in the master primal prob- lem. As a result, we can apply the dual decomposition to minimize independently the Lagrangian cost of all the MBs using the reference software model of H.264. Furthermore, the optimal Lagrange multi- plier λ∗ is iteratively derived from the solution of the dual problem. As an example, we derive the optimal bit allocation condition with the consideration of temporal prediction dependency among the pic- tures. Experimental results show that the proposed method achieves better performance than the reference software model of H.264 with rate control for given bit constraint.
2008 IEEE International Conference on Acoustics, Speech and Signal Processing, 2008
A variant of the k-nearest neighbor algorithm is proposed for image interpolation. Instead of usi... more A variant of the k-nearest neighbor algorithm is proposed for image interpolation. Instead of using a static volume or static k, the proposed algorithm determines a dynamic k that is small for inputs whose neighbors are very similar and large for inputs whose neighbors are dissimilar. Then, based on the neighbors that the adaptable k provides and their corresponding similarity measures, a weighted MMSE solution defines filters specific to intrinsic content of a low-resolution input image patch without yielding to the limitations of a non-uniformly distributed training set. Finally, global optimization through a single pass Markovian-like network further imposes on filter weights. The approach is justified by a sufficient quantity of relevant training pairs per test input and compared to current state of the art nearest neighbor interpolation techniques.
2010 IEEE International Conference on Image Processing, 2010
ABSTRACT We propose total subset variation (TSV), a convexity preserving generalization of the to... more ABSTRACT We propose total subset variation (TSV), a convexity preserving generalization of the total variation (TV) prior, for higher order clique MRF. A proposed differentiable approximation of the TSV prior makes it amenable for use in large images (e.g. 1080p). A convex relaxation of sub-exponential distribution is proposed as a criterion to determine the parameters of the optimization problem resulting from the TSV prior. For the super-resolution application, experiments show reconstruction error improvement with respect to the TV and other methods.
2006 International Conference on Image Processing, 2006
ABSTRACT
2006 IEEE Workshop on Multimedia Signal Processing, 2006
This paper proposes the application of learned kernels in support vector regression to superresol... more This paper proposes the application of learned kernels in support vector regression to superresolution in the discrete cosine transform (DCT) domain. Though previous works involve kernel learning, their problem formulation is examined to reformulate the semi-definite programming problem of finding the optimal kernel matrix. For the particular application to superresolution, downsampling properties derived in the DCT domain are exploited to
2006 IEEE Workshop on Multimedia Signal Processing, 2006
Abstract-In this paper, a novel motion compensated frame interpolation (MCFI) algorithm by mergin... more Abstract-In this paper, a novel motion compensated frame interpolation (MCFI) algorithm by merging blocks that have unreliable motion vectors (MVs) based on their residual errors is proposed. Unlike the conventional methods that find true motion using smaller blocks and vector ...
2004 International Conference on Image Processing, 2004. ICIP '04., 2004
In this paper, a ncw fast video codec using Wavelet Di@rence Reduction (WDR) algorirhm is present... more In this paper, a ncw fast video codec using Wavelet Di@rence Reduction (WDR) algorirhm is presented. This proposed video codec is inspired by the concept of motion JPEG: numelv. we adapted the eJf7cienr WDR still image compression ulgorirhm into U video compression ...
2004 International Conference on Image Processing, 2004. ICIP '04., 2004
ABSTRACT
2004 International Conference on Image Processing, 2004. ICIP '04., 2004
... DCT-BASED PHASE CORRELATION MOTION ESTIMATION Min Li, Mainak Biswas, Sarijeev Kumar and Truon... more ... DCT-BASED PHASE CORRELATION MOTION ESTIMATION Min Li, Mainak Biswas, Sarijeev Kumar and Truong Nguyen ... [71 J. Song and B:L. Yeo, A fast DCT-domain inverse motion compensation algorithm based on shared infor-mation in a microblock, Asilomur Conf, vol. ...
IEEE International Conference on Image Processing 2005, 2005
The design of a class of wavelet filters and its application in Scalable Video Coding (SVC) is di... more The design of a class of wavelet filters and its application in Scalable Video Coding (SVC) is discussed in detail in this paper. The design method uses maximal flat wavelet filters as prototype filters and incorporates all other wavelet filter design requirements. The designed wavelet filters are optimal in a sense that best tradeoff between high stopband attenuation of analysis lowpass filter H 0 (z) and flat passband response of synthesis lowpass filter F 0 (z) is achieved. The simulation that compares the performances of the designed filters and Daubechies (9,7) filters in SVC are illustrated.
2010 IEEE International Conference on Acoustics, Speech and Signal Processing, 2010
In standard digital color imaging, each pixel position acquires data for only one color plane and... more In standard digital color imaging, each pixel position acquires data for only one color plane and the remaining two color planes must be inferred through a process known as demosaicking. Furthermore, the image is susceptible to blurring artifacts due to a moving camera or fast moving subject. In this work we develop a robust framework to demosaick the color filter array (CFA) image while reducing the blur corrupting the image. We begin by defining a color motion blur model that describes the motion blur artifacts affecting color images. We then integrate the motion blur model in the demosaicking algorithm to obtain a computationally efficient framework for deblurring while demosaicking.
Proceedings of Third International Symposium on Time-Frequency and Time-Scale Analysis (TFTS-96), 1996
IEEE GLOBECOM 2007-2007 IEEE Global Telecommunications Conference, 2007
Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005., 2005
This paper describes a technique for delivering real-time video over low-bandwidth wireless chann... more This paper describes a technique for delivering real-time video over low-bandwidth wireless channels. The proposed approach transmits data at a fixed bit-rate specified by the encoder and provides an error correction mechanism for robust data transmission without the use of channel quality information. The performance of this technique in the presence of transmission errors is demonstrated.
Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005., 2005
A multirate filter bank model is considered for reconstruction of periodically sampled signals. I... more A multirate filter bank model is considered for reconstruction of periodically sampled signals. In contrast to many previous methods which considered perfect reconstruction of deterministic signals, this approach uses a known discrete-time cyclostationary signal model to find a minimum mean-squared error reconstruction solution. A primary advantage of this approach is that it does not require a minimum sampling density, allowing
2013 Fifth International Workshop on Quality of Multimedia Experience (QoMEX), 2013
ABSTRACT
2006 Fortieth Asilomar Conference on Signals, Systems and Computers, 2006
In this paper, we consider scalable video coding (SVC) which has higher complexity than H.264/AVC... more In this paper, we consider scalable video coding (SVC) which has higher complexity than H.264/AVC since it has spatial, temporal and quality scalability in addition to H.264/AVC functionality. Furthermore, inter-layer prediction and layered coding for spatial scalability make motion estimate and mode decision more complex. Therefore, we propose low complexity SVC schemes by using current developing SVC standard. It is archived by prediction method such as skip, direct, inter-layer MV prediction with fast mode and motion vector (MV) estimation at enhancement layer. In order to increase the performance of inter-layer MV prediction, combined MV interpolation is applied with adjustment of prediction direction. Additionally, fast mode and MV estimation are proposed from structural properties of motion-compensated temporal filtering (MCTF) to elaborate predicted macro block (MB) mode and MV. From the experi- mental results, proposed method has comparable performance to reference software model with significant lower complexity. I. INTRODUCTION
2008 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video, 2008
Two new attacks on multiview videos in the view space is investigated, which is also applicable t... more Two new attacks on multiview videos in the view space is investigated, which is also applicable to stereo and free-view television. The first attack involves the generation of new views from different viewpoints. The second attack involves change in the region of focus (the display plane or the zero disparity plane) in the multiview images. A scale invariant feature descriptor (SIFT) based fingerprinting mechanism which can identify such attacks is developed. An online verification system matches the SIFT descriptors of the original video (stored at a central database) to that of the attacked video. Results show that the method can detect such attacks well, and is useful for copy detection on the internet.
2008 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video, 2008
A procedure for obtaining high dynamic range (HDR) videos from multiple differently exposed image... more A procedure for obtaining high dynamic range (HDR) videos from multiple differently exposed image sequences from a camera array is explored. It is observed that using information along both viewspace (camera) and temporal axes results in a good estimate of the HDR image sequence. Images captured at longer exposures are subject to motion blur. A novel motion deblurring scheme is proposed, prior to the actual HDR mapping process. This involves a multiscale directional structure preservation procedure which uses information from adjacent views along camera-space and frames along time. The proposed deblurring scheme works in spite of illumination variations between images.
2007 IEEE International Conference on Image Processing, 2007
The proposed algorithm in this work provides superresolution for color images by using a learning... more The proposed algorithm in this work provides superresolution for color images by using a learning based technique that utilizes both generative and discriminant approaches. The combination of the two approaches is designed with a stochastic classification-regression frame- work where a color image patch is first classified by its content, and then, based on the class of the patch, a learned
2008 IEEE International Conference on Acoustics, Speech and Signal Processing, 2008
In this paper, we apply the primal-dual decomposition and subgradi- ent projection methods to sol... more In this paper, we apply the primal-dual decomposition and subgradi- ent projection methods to solve the rate-distortion optimization prob- lem with the constant bit rate constraint. The primal decomposition method enables spatial or temporal prediction dependency within a Group Of Picture (GOP) to be processed in the master primal prob- lem. As a result, we can apply the dual decomposition to minimize independently the Lagrangian cost of all the MBs using the reference software model of H.264. Furthermore, the optimal Lagrange multi- plier λ∗ is iteratively derived from the solution of the dual problem. As an example, we derive the optimal bit allocation condition with the consideration of temporal prediction dependency among the pic- tures. Experimental results show that the proposed method achieves better performance than the reference software model of H.264 with rate control for given bit constraint.
2008 IEEE International Conference on Acoustics, Speech and Signal Processing, 2008
A variant of the k-nearest neighbor algorithm is proposed for image interpolation. Instead of usi... more A variant of the k-nearest neighbor algorithm is proposed for image interpolation. Instead of using a static volume or static k, the proposed algorithm determines a dynamic k that is small for inputs whose neighbors are very similar and large for inputs whose neighbors are dissimilar. Then, based on the neighbors that the adaptable k provides and their corresponding similarity measures, a weighted MMSE solution defines filters specific to intrinsic content of a low-resolution input image patch without yielding to the limitations of a non-uniformly distributed training set. Finally, global optimization through a single pass Markovian-like network further imposes on filter weights. The approach is justified by a sufficient quantity of relevant training pairs per test input and compared to current state of the art nearest neighbor interpolation techniques.
2010 IEEE International Conference on Image Processing, 2010
ABSTRACT We propose total subset variation (TSV), a convexity preserving generalization of the to... more ABSTRACT We propose total subset variation (TSV), a convexity preserving generalization of the total variation (TV) prior, for higher order clique MRF. A proposed differentiable approximation of the TSV prior makes it amenable for use in large images (e.g. 1080p). A convex relaxation of sub-exponential distribution is proposed as a criterion to determine the parameters of the optimization problem resulting from the TSV prior. For the super-resolution application, experiments show reconstruction error improvement with respect to the TV and other methods.
2006 International Conference on Image Processing, 2006
ABSTRACT
2006 IEEE Workshop on Multimedia Signal Processing, 2006
This paper proposes the application of learned kernels in support vector regression to superresol... more This paper proposes the application of learned kernels in support vector regression to superresolution in the discrete cosine transform (DCT) domain. Though previous works involve kernel learning, their problem formulation is examined to reformulate the semi-definite programming problem of finding the optimal kernel matrix. For the particular application to superresolution, downsampling properties derived in the DCT domain are exploited to
2006 IEEE Workshop on Multimedia Signal Processing, 2006
Abstract-In this paper, a novel motion compensated frame interpolation (MCFI) algorithm by mergin... more Abstract-In this paper, a novel motion compensated frame interpolation (MCFI) algorithm by merging blocks that have unreliable motion vectors (MVs) based on their residual errors is proposed. Unlike the conventional methods that find true motion using smaller blocks and vector ...
2004 International Conference on Image Processing, 2004. ICIP '04., 2004
In this paper, a ncw fast video codec using Wavelet Di@rence Reduction (WDR) algorirhm is present... more In this paper, a ncw fast video codec using Wavelet Di@rence Reduction (WDR) algorirhm is presented. This proposed video codec is inspired by the concept of motion JPEG: numelv. we adapted the eJf7cienr WDR still image compression ulgorirhm into U video compression ...
2004 International Conference on Image Processing, 2004. ICIP '04., 2004
ABSTRACT
2004 International Conference on Image Processing, 2004. ICIP '04., 2004
... DCT-BASED PHASE CORRELATION MOTION ESTIMATION Min Li, Mainak Biswas, Sarijeev Kumar and Truon... more ... DCT-BASED PHASE CORRELATION MOTION ESTIMATION Min Li, Mainak Biswas, Sarijeev Kumar and Truong Nguyen ... [71 J. Song and B:L. Yeo, A fast DCT-domain inverse motion compensation algorithm based on shared infor-mation in a microblock, Asilomur Conf, vol. ...
IEEE International Conference on Image Processing 2005, 2005
The design of a class of wavelet filters and its application in Scalable Video Coding (SVC) is di... more The design of a class of wavelet filters and its application in Scalable Video Coding (SVC) is discussed in detail in this paper. The design method uses maximal flat wavelet filters as prototype filters and incorporates all other wavelet filter design requirements. The designed wavelet filters are optimal in a sense that best tradeoff between high stopband attenuation of analysis lowpass filter H 0 (z) and flat passband response of synthesis lowpass filter F 0 (z) is achieved. The simulation that compares the performances of the designed filters and Daubechies (9,7) filters in SVC are illustrated.
2010 IEEE International Conference on Acoustics, Speech and Signal Processing, 2010
In standard digital color imaging, each pixel position acquires data for only one color plane and... more In standard digital color imaging, each pixel position acquires data for only one color plane and the remaining two color planes must be inferred through a process known as demosaicking. Furthermore, the image is susceptible to blurring artifacts due to a moving camera or fast moving subject. In this work we develop a robust framework to demosaick the color filter array (CFA) image while reducing the blur corrupting the image. We begin by defining a color motion blur model that describes the motion blur artifacts affecting color images. We then integrate the motion blur model in the demosaicking algorithm to obtain a computationally efficient framework for deblurring while demosaicking.
Proceedings of Third International Symposium on Time-Frequency and Time-Scale Analysis (TFTS-96), 1996
IEEE GLOBECOM 2007-2007 IEEE Global Telecommunications Conference, 2007
Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005., 2005
This paper describes a technique for delivering real-time video over low-bandwidth wireless chann... more This paper describes a technique for delivering real-time video over low-bandwidth wireless channels. The proposed approach transmits data at a fixed bit-rate specified by the encoder and provides an error correction mechanism for robust data transmission without the use of channel quality information. The performance of this technique in the presence of transmission errors is demonstrated.
Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005., 2005
A multirate filter bank model is considered for reconstruction of periodically sampled signals. I... more A multirate filter bank model is considered for reconstruction of periodically sampled signals. In contrast to many previous methods which considered perfect reconstruction of deterministic signals, this approach uses a known discrete-time cyclostationary signal model to find a minimum mean-squared error reconstruction solution. A primary advantage of this approach is that it does not require a minimum sampling density, allowing
2013 Fifth International Workshop on Quality of Multimedia Experience (QoMEX), 2013
ABSTRACT
2006 Fortieth Asilomar Conference on Signals, Systems and Computers, 2006
In this paper, we consider scalable video coding (SVC) which has higher complexity than H.264/AVC... more In this paper, we consider scalable video coding (SVC) which has higher complexity than H.264/AVC since it has spatial, temporal and quality scalability in addition to H.264/AVC functionality. Furthermore, inter-layer prediction and layered coding for spatial scalability make motion estimate and mode decision more complex. Therefore, we propose low complexity SVC schemes by using current developing SVC standard. It is archived by prediction method such as skip, direct, inter-layer MV prediction with fast mode and motion vector (MV) estimation at enhancement layer. In order to increase the performance of inter-layer MV prediction, combined MV interpolation is applied with adjustment of prediction direction. Additionally, fast mode and MV estimation are proposed from structural properties of motion-compensated temporal filtering (MCTF) to elaborate predicted macro block (MB) mode and MV. From the experi- mental results, proposed method has comparable performance to reference software model with significant lower complexity. I. INTRODUCTION