Muid Mufti

Fuzzy case-based reasoning for facial expression recognition

Fuzzy logic (FL) and case-based reasoning (CBR) are two well-known techniques for the implementation of intelligent classification systems. Each technique has its own advantages and drawbacks. FL, for example, provides an intuitive user interface, simplifies the process of knowledge representation, and minimizes the systems computational complexity in terms of time and memory usage. On the other hand, FL has problems in knowledge elicitation which render it difficult to adopt for intelligent system implementation. CBR avoids these problems by making use of past input-output data to decide the system output for the present input. The accuracy of CBR system grows as the number of cases increase. However, more cases can mean added computational complexity in terms of space and time. In this paper we make the proposition that a hybrid system comprising a blend of FL and CBR can lead to a solution where the two approaches cover each others weaknesses and benefit from each others strengths. We support our claim by taking the problem of facial expression recognition from an input image. The facial expression recognition system presented in this paper uses a case base populated with fuzzy rules for recognizing each expression. Experimental results demonstrate that the system inherits the strengths of both methods.

Fuzzy Rule Based Facial Expression Recognition

Facial expressions are a key modality in human communication. Facial expressions occur simultaneously with social interaction and help in interpretation of spoken communication. The understanding of facial expressions is a basic requirement in the development of next generation human computer interaction (HCI) systems. Facial expressions, however, are qualitative in nature and it is very difficult to explain them by a mathematical model. This paper presents a system framework that uses fuzzy logic principles to recognize facial expressions in video sequences. The system uses image processing techniques to detect MPEG4 facial animation parameters (FAPs) in an input video. A fuzzy rule base has been developed to interpret the collective set of FAPs as constituting a particular facial expression.

Real time mono vision gesture based virtual keyboard system

This paper presents a novel mono-vision virtual keyboard design for consumers of mobile and portable computing devices such as PDAs, mobile phones etc. Fuzzy approaches to gesture recognition are developed to reveal the key pressed over the printed sheet keyboard by analyzing the hand and finger gesture captured in the video sequence. Real time system is developed by integrating SDIO camera with PDA in the application environment. Reliable results are experienced by the implementation of the proposed real time mono vision gestured virtual keyboard system

Intelligent Expression Blending for Performance Driven Facial Animation

Performance-Driven Facial Animation (PDFA) allows an animator to specify the desired facial action in terms of detected face movements of a performer. The animation system works by mimicking the actions of the performer on an animated character. Research efforts in PDFA have mostly concentrated on efficient tracking and re-targeting of facial expressions; very little attention has been paid to the intelligent enhancement of expressions. This paper describes an intelligent system that automatically enhances facial expressions in a PDFA sequence in light of its context. The system models facial expressions as MPEG-4 compliant fuzzy streams. The video context is processed by a fuzzy rule based system (FRBS) in order to find out what expression needs to be added to the detected expression to enhance its naturalness. The FRBS then intelligently blends the detected expression and the expression implied by the context to generate the enhanced expression parameters which are used to animate the target 3D facial model.

On Modeling Post Decryption Error Processes in UMTS Air Interface

This paper analyzes the impact of encryption over the UMTS air interface. Using a finite state Markov characterization of the UMTS decryption process, a stochastic model has been developed that quantifies the impact of bit errors in the ciphertext and cipher synchronization counter. The effects of residual bit errors, UMTS air interface power budget, interleaving span and channel coding rate on the decryption process are analyzed.

A novel compression technique for multi-camera nodes through directional correlation

A major fraction of multimedia stream contents tends to be redundant and leads to wastage of storage capacity and channel bandwidth. In order to eliminate surplus data, standard video compression algorithms exploit spatial and temporal correlation present in video sequence. However, in case of a multisensor network, intersensor statistical redundancy is the most significant factor in acquiring efficient link utilization as well as making resultant findings valuable to the end user. In this paper, an extension to our previously proposed scheme has been presented to accomplish performance goals of a multisensor environment. Standard MPEG codec has been used to accomplish distributed motion compensation in prespecified directions known as directional correlation. Video frame correlation has been estimated locally at the camera node as well as across different nodes, defined as node communication strategies. Further, receiver feedback assists in quality control after reconstitution by decoder assessment. Results estimated have been analyzed for saving ratios and multimedia quality. Results analysis illustrates increased gains in frame quality and compression saving, achieved through reducing node displacement from the reference node N R .

A heuristically guided hybrid approach to face detection for content based image retrieval in internet images

Content based image retrieval (CBIR) from the Internet often requires human face detection against a cluttered background. In this paper, a solution to this problem using colored images is presented. The method combines human skin color detection with knowledge based heuristics for efficient and robust face detection. Skin colored clusters are extracted from the image using color distribution of human skin. The reduced data set is processed by neural networks to detect various facial feature candidates. Common knowledge based heuristics are applied to the detected facial feature candidates. Face is detected only when both the color and features match the human face. The proposed method, which can detect multiple faces in an image, is invariant to scale and position.

A bi-level hierarchical fuzzy object based approach to natural scene classification

A fuzzy logic based approach for modeling various categories of natural scenes in images is presented. Different local objects occurring in a natural scene are represented as fuzzy semantic objects having various color and texture attributes. The entire scene is viewed as a two-level abstraction hierarchy. At the top level, the scene is considered to be comprised of three separate regions defined across its vertical dimension. At the lower level, each region is viewed as a collection of several local fuzzy objects. The frequency of occurrence of fuzzy objects in various vertical regions is used to determine the scene class. The system has been tested on a standard dataset and shows good results.