Articles de revista
http://hdl.handle.net/2117/3333
Sat, 10 Dec 2016 13:17:18 GMT2016-12-10T13:17:18ZObject-based image coding
http://hdl.handle.net/2117/97957
Object-based image coding
Pardàs Feliu, Montse
Object-based compression methods describe images in terms of a set of regions (a partition), and of some information for each region to be used by the receiver to reconstruct the image (its texture). Different techniques can be used to define the partition as well as for coding it and its texture. In this paper we propose a general multi-resolution segmentation algorithm which can deal with many different types of images selecting the appropriated criteria. We also review most outstanding techniques for coding in this context.
Fri, 09 Dec 2016 16:07:53 GMThttp://hdl.handle.net/2117/979572016-12-09T16:07:53ZPardàs Feliu, MontseObject-based compression methods describe images in terms of a set of regions (a partition), and of some information for each region to be used by the receiver to reconstruct the image (its texture). Different techniques can be used to define the partition as well as for coding it and its texture. In this paper we propose a general multi-resolution segmentation algorithm which can deal with many different types of images selecting the appropriated criteria. We also review most outstanding techniques for coding in this context.Segmentation of video sequences for partition tree generation
http://hdl.handle.net/2117/97946
Segmentation of video sequences for partition tree generation
Pardàs Feliu, Montse; Salembier Clairon, Philippe Jean
In this paper we propose a segmentation structure to generate a partition tree with multiple partition proposals. These multiple partitions are obtained for every image in a hierarchical way, from the coarsest one to the finest one. Time evolution of the regions is defined through a projection step which relates the previous coded partition with the current partition tree. Finer partitions than the projected one are defined segmenting the projected partition using texture criteria, while coarser partitions are obtained merging regions with motion criteria. Morphological tools are used both in the projection and the re-segmentation steps.
Fri, 09 Dec 2016 14:05:11 GMThttp://hdl.handle.net/2117/979462016-12-09T14:05:11ZPardàs Feliu, MontseSalembier Clairon, Philippe JeanIn this paper we propose a segmentation structure to generate a partition tree with multiple partition proposals. These multiple partitions are obtained for every image in a hierarchical way, from the coarsest one to the finest one. Time evolution of the regions is defined through a projection step which relates the previous coded partition with the current partition tree. Finer partitions than the projected one are defined segmenting the projected partition using texture criteria, while coarser partitions are obtained merging regions with motion criteria. Morphological tools are used both in the projection and the re-segmentation steps.Morphological tools for segmentation: connected operators and watersheds
http://hdl.handle.net/2117/97945
Morphological tools for segmentation: connected operators and watersheds
Meyer, F; Oliveras Vergés, Albert; Salembier Clairon, Philippe Jean; Vachier, C
The paper makes a survey of the morphological tools which are useful for image simplification and segmentation. It starts presenting the connected filters, which have the unique feature to simplify the images without blurring the contours of the remaining objects. Such filters may be constructed on several criteria : size, contrast, complexity or type of motion. In the case where geometric criteria are used, it is possible to base a scale-space description of the images on a series of increasing filters. This is most useful when applied to a gradient image : successive filters will progressively fill the valleys of the gradient image and a watershed transform applied on these images will produce segmentations which are coarser and coarser. These segmentations have a tree structure : from a finer level to a coarser level, only fusions take place.
Fri, 09 Dec 2016 13:59:28 GMThttp://hdl.handle.net/2117/979452016-12-09T13:59:28ZMeyer, FOliveras Vergés, AlbertSalembier Clairon, Philippe JeanVachier, CThe paper makes a survey of the morphological tools which are useful for image simplification and segmentation. It starts presenting the connected filters, which have the unique feature to simplify the images without blurring the contours of the remaining objects. Such filters may be constructed on several criteria : size, contrast, complexity or type of motion. In the case where geometric criteria are used, it is possible to base a scale-space description of the images on a series of increasing filters. This is most useful when applied to a gradient image : successive filters will progressively fill the valleys of the gradient image and a watershed transform applied on these images will produce segmentations which are coarser and coarser. These segmentations have a tree structure : from a finer level to a coarser level, only fusions take place.General requirements for coding oriented segmentation of video-sequences
http://hdl.handle.net/2117/97944
General requirements for coding oriented segmentation of video-sequences
Marqués Acosta, Fernando; Pardàs Feliu, Montse; Salembier Clairon, Philippe Jean; Meyer, F
This paper deals with the basic requirements that should be fulfilledby a technique for segmenting video sequences in coding applications. The specific problems of coding-oriented video segmentation are analyzed. This way intra-frame and inter-frame segmentation approaches are studied. In the inter-frame mode, the problems of temporal label coherence and connectivity in the coding framework are discussed and solutions are presented.
Fri, 09 Dec 2016 13:50:01 GMThttp://hdl.handle.net/2117/979442016-12-09T13:50:01ZMarqués Acosta, FernandoPardàs Feliu, MontseSalembier Clairon, Philippe JeanMeyer, FThis paper deals with the basic requirements that should be fulfilledby a technique for segmenting video sequences in coding applications. The specific problems of coding-oriented video segmentation are analyzed. This way intra-frame and inter-frame segmentation approaches are studied. In the inter-frame mode, the problems of temporal label coherence and connectivity in the coding framework are discussed and solutions are presented.A region-based subband coding scheme
http://hdl.handle.net/2117/97938
A region-based subband coding scheme
Casas Pla, Josep Ramon; Torres Urgell, Lluís
This paper describes a region-based subband coding scheme intended for efficient representation of the visual information contained in image regions of arbitrary shape. QMF filters are separately applied inside each region for the analysis and synthesis stages, using a signal-adaptive symmetric extension technique at region borders. The frequency coefficients corresponding to each region are identified over the various subbands of the decomposition, so that the coding steps — namely, bit-allocation, quantization and entropy coding — can be performed independently for each region.
Region-based subband coding exploits the possible homogeneity of the region contents by distributing the available bitrate not only in the frequency domain but also in the spatial domain, i.e. among the considered regions. The number of bits assigned to the subbands is optimized region by region for the whole image, by means of a rate-distortion optimization algorithm. Improved compression efficiency is obtained thanks to the local adaptativity of the bit allocation to the spectral contents of the different regions. This compensates for the overhead data spent in the coding of contour information.
As the subband coefficients obtained for each region are coded as separate data units, the content-based functionalities required for the future MPEG4 video coding standard can be readily handled. For instance, content-based scalability is possible by simply imposing user-defined constraints to the bit-assignment in some regions.
Fri, 09 Dec 2016 13:22:45 GMThttp://hdl.handle.net/2117/979382016-12-09T13:22:45ZCasas Pla, Josep RamonTorres Urgell, LluísThis paper describes a region-based subband coding scheme intended for efficient representation of the visual information contained in image regions of arbitrary shape. QMF filters are separately applied inside each region for the analysis and synthesis stages, using a signal-adaptive symmetric extension technique at region borders. The frequency coefficients corresponding to each region are identified over the various subbands of the decomposition, so that the coding steps — namely, bit-allocation, quantization and entropy coding — can be performed independently for each region.
Region-based subband coding exploits the possible homogeneity of the region contents by distributing the available bitrate not only in the frequency domain but also in the spatial domain, i.e. among the considered regions. The number of bits assigned to the subbands is optimized region by region for the whole image, by means of a rate-distortion optimization algorithm. Improved compression efficiency is obtained thanks to the local adaptativity of the bit allocation to the spectral contents of the different regions. This compensates for the overhead data spent in the coding of contour information.
As the subband coefficients obtained for each region are coded as separate data units, the content-based functionalities required for the future MPEG4 video coding standard can be readily handled. For instance, content-based scalability is possible by simply imposing user-defined constraints to the bit-assignment in some regions.Stochastic vector quantization of images
http://hdl.handle.net/2117/97894
Stochastic vector quantization of images
Torres Urgell, Lluís; Casas Pla, Josep Ramon; Arias, E
One of the most important steps in the vector quantization of images is the design of the codebook. The codebook is generally designed using the LBG algorithm, that is in essence a clustering algorithm which uses a large training set of empirical data that is statistically representative of the image to be quantized. The LBG algorithm, although quite effective for practical applications, is computationally very expensive and the resulting codebook has to be recalculated each time the type of image to be encoded changes. One alternative to the generation of the codebook, called stochastic vector quantization, is presented in this paper. Stochastic vector quantization (SVQ) is based on the generation of the codebook according to some previous model defined for the image to be encoded. The well-known AR model has been used to model the image in the current implementations of the technique, and has shown good performance in the overall scheme. To show the merit of the technique in different contexts, stochastic vector quantization is discussed and applied to both pixel-based and segmentation-based image coding schemes.
Wed, 07 Dec 2016 16:28:13 GMThttp://hdl.handle.net/2117/978942016-12-07T16:28:13ZTorres Urgell, LluísCasas Pla, Josep RamonArias, EOne of the most important steps in the vector quantization of images is the design of the codebook. The codebook is generally designed using the LBG algorithm, that is in essence a clustering algorithm which uses a large training set of empirical data that is statistically representative of the image to be quantized. The LBG algorithm, although quite effective for practical applications, is computationally very expensive and the resulting codebook has to be recalculated each time the type of image to be encoded changes. One alternative to the generation of the codebook, called stochastic vector quantization, is presented in this paper. Stochastic vector quantization (SVQ) is based on the generation of the codebook according to some previous model defined for the image to be encoded. The well-known AR model has been used to model the image in the current implementations of the technique, and has shown good performance in the overall scheme. To show the merit of the technique in different contexts, stochastic vector quantization is discussed and applied to both pixel-based and segmentation-based image coding schemes.Motion estimation using higher-order statistics
http://hdl.handle.net/2117/97801
Motion estimation using higher-order statistics
Sayrol Clols, Elisa; Gasull Llampallas, Antoni; Rodríguez Fonollosa, Javier
The objective of this paper is to introduce a fourth-order cost function of the displaced frame difference (DFD) capable of estimating motion even for small regions or blocks. Using higher than second-order statistics is appropriate in case the image sequence is severely corrupted by additive Gaussian noise. Some results are presented and compared to those obtained from the mean kurtosis and the mean square error of the DFD.
Mon, 05 Dec 2016 16:29:03 GMThttp://hdl.handle.net/2117/978012016-12-05T16:29:03ZSayrol Clols, ElisaGasull Llampallas, AntoniRodríguez Fonollosa, JavierThe objective of this paper is to introduce a fourth-order cost function of the displaced frame difference (DFD) capable of estimating motion even for small regions or blocks. Using higher than second-order statistics is appropriate in case the image sequence is severely corrupted by additive Gaussian noise. Some results are presented and compared to those obtained from the mean kurtosis and the mean square error of the DFD.Morphological operators for image and video compression
http://hdl.handle.net/2117/97797
Morphological operators for image and video compression
Salembier Clairon, Philippe Jean; Brigger, P; Casas Pla, Josep Ramon; Pardàs Feliu, Montse
This paper deals with the use of some morphological tools for image and video coding. Mathematical morphology can be considered as a shape-oriented approach to signal processing, and some of its features make it very useful for compression. Rather than describing a coding algorithm, the purpose of this paper is to describe some morphological tools that have proved attractive for compression. Four sets of morphological transformations are presented: connected operators, the region-growing version of the watershed, the geodesic skeleton, and a morphological interpolation technique. The authors discuss their implementation, and show how they can be used for image and video segmentation, contour coding, and texture coding.
Mon, 05 Dec 2016 16:18:52 GMThttp://hdl.handle.net/2117/977972016-12-05T16:18:52ZSalembier Clairon, Philippe JeanBrigger, PCasas Pla, Josep RamonPardàs Feliu, MontseThis paper deals with the use of some morphological tools for image and video coding. Mathematical morphology can be considered as a shape-oriented approach to signal processing, and some of its features make it very useful for compression. Rather than describing a coding algorithm, the purpose of this paper is to describe some morphological tools that have proved attractive for compression. Four sets of morphological transformations are presented: connected operators, the region-growing version of the watershed, the geodesic skeleton, and a morphological interpolation technique. The authors discuss their implementation, and show how they can be used for image and video segmentation, contour coding, and texture coding.Flat zones filtering, connected operators, and filters by reconstruction
http://hdl.handle.net/2117/97786
Flat zones filtering, connected operators, and filters by reconstruction
Salembier Clairon, Philippe Jean; Serra i Aguadé, Josep
This correspondence deals with the notion of connected operators. Starting from the definition for operator acting on sets, it is shown how to extend it to operators acting on function. Typically, a connected operator acting on a function is a transformation that enlarges the partition of the space created by the flat zones of the functions. It is shown that from any connected operator acting on sets, one can construct a connected operator for functions (however, it is not the unique way of generating connected operators for functions). Moreover, the concept of pyramid is introduced in a formal way. It is shown that, if a pyramid is based on connected operators, the flat zones of the functions increase with the level of the pyramid. In other words, the flat zones are nested. Filters by reconstruction are defined and their main properties are presented. Finally, some examples of application of connected operators and use of flat zones are described.
Mon, 05 Dec 2016 14:30:17 GMThttp://hdl.handle.net/2117/977862016-12-05T14:30:17ZSalembier Clairon, Philippe JeanSerra i Aguadé, JosepThis correspondence deals with the notion of connected operators. Starting from the definition for operator acting on sets, it is shown how to extend it to operators acting on function. Typically, a connected operator acting on a function is a transformation that enlarges the partition of the space created by the flat zones of the functions. It is shown that from any connected operator acting on sets, one can construct a connected operator for functions (however, it is not the unique way of generating connected operators for functions). Moreover, the concept of pyramid is introduced in a formal way. It is shown that, if a pyramid is based on connected operators, the flat zones of the functions increase with the level of the pyramid. In other words, the flat zones are nested. Filters by reconstruction are defined and their main properties are presented. Finally, some examples of application of connected operators and use of flat zones are described.Region-based video coding using mathematical morphology
http://hdl.handle.net/2117/97723
Region-based video coding using mathematical morphology
Salembier Clairon, Philippe Jean; Torres Urgell, Lluís; Meyer, F; Gu, C
This paper presents a region-based coding algorithm for video sequences. The coding approach involves a time-recursive segmentation relying on the pixels homogeneity, a region-based motion estimation, and motion compensated contour and texture coding. This algorithm is mainly devoted to very low bit rate video coding applications. One of the important features of the approach is that no assumption is made about the sequence content. Moreover the algorithm structure leads to a scalable coding process giving various levels of quality and bit rates. The coding as well as the segmentation are controlled to regulate the bit stream. Finally, the interest of morphological tools in the content of region-based coding is extensively reviewed.
Fri, 02 Dec 2016 16:55:49 GMThttp://hdl.handle.net/2117/977232016-12-02T16:55:49ZSalembier Clairon, Philippe JeanTorres Urgell, LluísMeyer, FGu, CThis paper presents a region-based coding algorithm for video sequences. The coding approach involves a time-recursive segmentation relying on the pixels homogeneity, a region-based motion estimation, and motion compensated contour and texture coding. This algorithm is mainly devoted to very low bit rate video coding applications. One of the important features of the approach is that no assumption is made about the sequence content. Moreover the algorithm structure leads to a scalable coding process giving various levels of quality and bit rates. The coding as well as the segmentation are controlled to regulate the bit stream. Finally, the interest of morphological tools in the content of region-based coding is extensively reviewed.