Papers by Somnath Sengupta
… Computer Networks …, Jan 1, 2009
The IEEE 802.16 standard specifies a MeSH mode of operation which permits the setup of Wireless M... more The IEEE 802.16 standard specifies a MeSH mode of operation which permits the setup of Wireless Mesh Networks (WMN) with per-link QoS support. The standard specifies both distributed as well as centralized reservation schemes. Distributed scheduling is highly flexible, and enables operation of the WMN even in the absence of a central controlling instance or base station. A systematic study of strategies for distributed scheduling in the IEEE 802.16 MeSH mode is, however, missing. In this paper we model the individual links in the 802.16 WMN and design and derive efficient strategies for distributed scheduling to reserve bandwidth required for transmission on the modelled link. Additionally, we evaluate our proposed reservation model using simulations, study the impact of key parameters and identify issues for further research in WiMAX based WMNs. • nrtPS traffic: the network traffic trace file BC-pAug89.TL (around 1.4016 Mbps) (source see [7]).
Multimedia Technology …, Jan 1, 2011
Abstract Conventional fast search motion estimation algorithms use a heuristically based approach... more Abstract Conventional fast search motion estimation algorithms use a heuristically based approach to determine their search window. By making use of temporal correlation between the current frame and the past frame and spatial correlation between the neighborhood ...
Signal and Image Processing …, Jan 1, 2011
Abstract In this paper, we examine the suitability of correlogram for background subtraction, as ... more Abstract In this paper, we examine the suitability of correlogram for background subtraction, as a step towards moving object detection. Correlogram captures inter-pixel relationships in a region and is seen to be effective for modelling dynamic backgrounds. We propose ...
… National Conference on Communications (February 01 …, Jan 1, 2008
… , 2011 IEEE Pacific Rim Conference on, Jan 1, 2011
Error Concealment (EC) techniques attempt to reconstruct the lost area of a frame in a video sequ... more Error Concealment (EC) techniques attempt to reconstruct the lost area of a frame in a video sequence using spatial and/or temporal correlation. Temporal error concealment (TEC) performs better than the spatial error concealment (SEC) in terms of PSNR of the reconstructed macroblocks. However, the performance of TEC deteriorates drastically in presence of scene transitions due to lack of temporal correlation. In this paper, we propose an encoder driven scene transition detection which would facilitate error resilience at the encoder and concealment at the decoder. A new edge-direction based spatial error concealment scheme termed DEBSEC is also proposed.
Society of Photo-Optical …, Jan 1, 2012
ABSTRACT
cmlab.csie.ntu.edu.tw
This paper proposes a unidirectional encoder rate control (ERC) scheme in the interpolation based... more This paper proposes a unidirectional encoder rate control (ERC) scheme in the interpolation based distributed video coding. As the encoder is complexity constrained, accurate estimation of number of bits to decode each bit plane is indeed difficult at the encoder. In case of under-estimation of the bits, correction of the errors in the decoded bit planes, by utilizing the available information, is one of the important tasks at the decoder. This was addressed by recent schemes. In this paper, we present an improved ERC, considering higher group of pictures(GOP). The contributions of the proposed scheme are (1) adaptive rate estimation, considering the dependency across Wyner-Ziv frames (2) motion adaptive reconstruction and (3) Side information refinement after decoding all the frames in the GOP. The proposed scheme is tested with several sequences, showing improvements in the case of GOP-4.
cmlab.csie.ntu.edu.tw
Distributed video coding (DVC) is an emerging coding paradigm, aiming at low complexity encoders.... more Distributed video coding (DVC) is an emerging coding paradigm, aiming at low complexity encoders. This paper proposes a new scheme for transform domain Wyner-Ziv (WZ) video codec, where the key and WZ frames are encoded in multiple layers. The layers of the each frame are generated by sub sampling the 4x4 blocks in the spatial domain. After decoding each layer, side information (SI) refinement process is employed to improve the quality of SI to decode the next WZ layer. The codec performance is further improved by the decoder-driven adaptive skip/WZ control, estimated using the refined SI and the correlation noise. Rate-Distortion performance of the proposed scheme is tested with several sequences and performance improvements are noted.
Electronics letters, Jan 1, 2012
ABSTRACT Proposed is a post-processing scheme to reduce the quantisation noise of the key frames ... more ABSTRACT Proposed is a post-processing scheme to reduce the quantisation noise of the key frames by using the decoded neighbour Wyner-Ziv frames in a distributed video coding framework. The proposed scheme results in a quality gain of up to 2 dB for the key frames and improves the overall rate-distortion performance of the codec without adding any additional complexity at the encoder.
Journal of Visual Communication and Image …, Jan 1, 2012
Moving object detection in dynamic backgrounds remains a challenging problem. Our earlier work es... more Moving object detection in dynamic backgrounds remains a challenging problem. Our earlier work established that the background subtraction using the covariance matrix descriptor is robust for dynamic backgrounds. The work proposed herein extends this approach further, using just two features-Hu moment and intensity. An improved local Hu moment is proposed, where the moment calculation of a pixel, involving neighboring pixels, are used in a weighted manner to reduce the effects of background moving pixels and the accurate shape localization of moving objects simultaneously. To further counter the erratic labeling of dynamic pixels, the fact that the neighboring pixels are spatially correlated is exploited for model construction and foreground detection. An adaptive model updating rate is calculated as a function of model distance. The proposed approach models each pixel with a covariance matrix and a mean feature vector and is dynamically updated. Extensive studies are made with the proposed technique to demonstrate its effectiveness.
Image and Vision Computing, Jan 1, 2012
Background subtraction Rough set 3D histon 3D fuzzy histon 3D HRI 3D FHRI Detection of moving obj... more Background subtraction Rough set 3D histon 3D fuzzy histon 3D HRI 3D FHRI Detection of moving objects in the presence of challenging background situations like swaying vegetation, rippling water, camera jitter etc., is known to be a difficult task. Background subtraction is considered to be better than the other approaches in terms of robustness. Its success primarily depends on the proper choice of background model(s) associated with every pixel for its foreground/background labeling. In this work, we have adopted rough-set theoretic measures to embed the spatial similarity around a neighborhood as a model for the pixel. Basic histon and its associated measure Basic Histon Roughness Index (BHRI) have been reported in the literature. It was applied to still image segmentation with impressive performance. Its adoption in video sequences for foreground/background labeling is proposed herein. We extended the histon concept to a 3D histon, which considers the intensities across the color planes in a combined manner, instead of considering independent color planes. Further, we also incorporated fuzziness into the 3D HRI measure. The labeling decision is based on Bhattacharyya distance between the model HRI and the corresponding measure in the current frame. Adoption of rough set theoretic concept into moving object segmentation is nontrivial, as the model updating requires careful consideration so that the pixels associated with gradually changing background or dynamic background are labeled as background and at the same time, slow moving objects are never adopted into the background model. A novel background model update strategy proposed herein takes these into consideration and also eliminates the need of having exclusive ideal background frame initially.
IEEE SIGNAL PROCESSING …, Jan 1, 2012
Robust detection of moving objects in presence of dynamic backgrounds is yet a challenging proble... more Robust detection of moving objects in presence of dynamic backgrounds is yet a challenging problem. In this letter, we propose a fuzzy membership transformation to be applied on the co-occurrence vector to derive a rich fuzzy transformed co-occurrence vector with shared membership values in a reduced dimensionality vector space. Fuzzy statistical texture features, derived from this fuzzy transformed co-occurrence vector, are able to improve the robustness in detecting moving objects, as compared to the traditional statistical texture features and other contemporary moving object segmentation approaches.
Pattern Recognition Letters, Jan 1, 2001
In this paper, we propose a genetic algorithm (GA)-based approach to determine the external param... more In this paper, we propose a genetic algorithm (GA)-based approach to determine the external parameters of the camera from the knowledge of a given set of points in object space. We study the eect of noise and presence of outliers, and also mismatch resulting from incorrect correspondences between the object space points and the image space points, on the estimation of three translation parameters and three rotational parameters of a camera. The average of the magnitudes of the translation errors varies from 2.25 cm to 5 mm and the average of the magnitudes of the rotational errors varies from 0.4°to 0.25°at 20 dB SNR. The error in parameter estimation is insigni®cant upto three pairs of mismatched points out of 20 points in object space and skyrockets when four or more pairs of points are mismatched. These results have clearly established the robustness of GA in external camera parameter estimation. Ó
Computer VisionACCV 2006, Jan 1, 2006
Abstract. A five layered, event driven hierarchical framework for generic sports video classifica... more Abstract. A five layered, event driven hierarchical framework for generic sports video classification has been proposed in this paper. The top layer classifications are based on a few popular audio and video con-tent analysis techniques like short-time energy and Zero Crossing Rate ...
Image Processing, 2003. …, Jan 1, 2003
ABSTRACT This work presents a new perceptually motivated bit allocation strategy for Region of In... more ABSTRACT This work presents a new perceptually motivated bit allocation strategy for Region of Interest (ROI) coding in video sequences. It is not possible to assure any minimum target foreground quality in the approaches reported so far in the literature. In our proposed strategy ...
2006 IEEE International Conference on …, Jan 1, 2006
In this paper, we present a novel approach towards customized and automated generation of sports ... more In this paper, we present a novel approach towards customized and automated generation of sports highlights from its extracted events and semantic concepts. A recorded sports video is first divided into slots, based on the game progress and for each slot, an importance-based concept and event-selection is proposed to include those in the highlights. Using our approach, we have successfully extracted highlights from recorded video of cricket match.
Journal of …, Jan 1, 2009
Pattern Recognition, Jan 1, 2004
This paper describes a novel and fast approach to Full Search Block Matching (FSBM) employing pre... more This paper describes a novel and fast approach to Full Search Block Matching (FSBM) employing prediction of search region, based on the sampled statistics of Sum-of-Absolute Difference (SAD) distributions. The motion vector is predicted to belong to either of two regions, ...
Multimedia Tools and Applications, Jan 1, 2010
This paper presents a novel approach towards automated highlight generation of broadcast sports v... more This paper presents a novel approach towards automated highlight generation of broadcast sports video sequences from its extracted events and semantic concepts. A sports video is hierarchically divided into temporal partitions namely, megaslots, slots, and semantic entities, namely concepts, and events. The proposed method extracts event sequence from video and classifies each sequence into a concept by sequential association mining. The extracted concepts and events within the concepts are selected according to their degree of importance to include those in the highlights. A parameter degree of abstraction is proposed, which gives a choice to the user about how concisely the extracted concepts should be produced for a specified highlight duration. We have successfully extracted highlights from recorded video of cricket match and compared our results with the manually-generated highlights by sports television channel.
… Workshop on Very Low Bitrate Video …, Jan 1, 2001
In this paper, we propose modifications to a H.263 codec for coding very low bit-rate head and sh... more In this paper, we propose modifications to a H.263 codec for coding very low bit-rate head and shoulder video sequences. The important facial features (eyes, nose and mouth) are tracked from frame to frame and the knowledge of their positions is used to improve quantization in these regions of interest, compared to the other facial regions and the background, which have coarse quantization. Results are presented which show the effectiveness of the method using common head and shoulder video sequences.
Uploads
Papers by Somnath Sengupta