Zhou Chang-le

A novel face recognition method with feature combination

A novel combined personalized feature framework is proposed for face recognition (FR). In the framework, the proposed linear discriminant analysis (LDA) makes use of the null space of the within-class scatter matrix effectively, and Global feature vectors (PCA-transformed) and local feature vectors (Gabor wavelet-transformed) are integrated by complex vectors as input feature of improved LDA. The proposed method is compared to other commonly used FR methods on two face databases (ORL and UMIST). Results demonstrated that the performance of the proposed method is superior to that of traditional FR approaches

Real-time non-intrusive eye tracking for human-computer interaction

A real-time eye tracking method was introduced. The method integrates several algorithms such as AdaBoost, optical flow, and Camshift. First, lighting condition was standardized, and the camera was placed below face. Second, the nostril tracking points were localized and taken as reference points. Third, the Camshift algorithm and Lucas-Kanade optical flow algorithm were respectively utilized to track the face and nostrils, and the pupils were positioned using gradient Hough circle transform. Finally, with the coordinates of nostrils and pupils, the velocity and trajectory of eye movement were calculated. The experiments on a video of resolution at 640⋆480 result a tracking speed of 25 frames per second. Results prove the effectiveness and efficiency of this method.

Study on complexion recognition in TCM

Based on complexion-viscera diagram in TCM (Traditional Chinese Medicine), a method for complexion recognition was proposed for the first time. Firstly, the sampling environment was standardized, and 68 common facial feature points were localized by AdaBoost and ASM algorithm. Second, the complexion features in LAB color space were extracted from 15 diagnostic feature points on complexion-viscera diagram by FCM clustering which separates the complexion feature from skin color. Finally, the complexion features were classified by SVM algorithm. This method can be applied to the automatic logic inference of face diagnosis in TCM. Experiments on a test set of 400 face images result an accuracy of 84.6%. Results prove the effectiveness and efficiency of this method.

Research on Chinese ellipsis recovering based on discourse representation theory

This paper presents our research work on Chinese ellipsis recovering based on discourse representation theory (DRT). From various perspectives, our work focuses on how to construct the Chinese DRT and the semantic model. Then it proposes an ellipsis recovering model based on discourse, which selects the most suitable one and fill in the syntactic ellipsis and possible semantic ellipsis. The simulated results show some advantages of the proposed approach based on DRT and potential applications in Chinese oral system and discourse machine understanding

A hierarchical approach to simulation of the melodic harmonization process

In this paper, a hybrid system (method) capable of generating accompaniment to human-created melodies was proposed with the use of machine learning and theoretical knowledge, as well as the idea of songs global framework to simulate the real composition process. This paper introduced a new method in which the harmonization process were divided into several hierarchies, chords in every hierarchy were expanded from their father hierarchy and were the highest several ranked ones among candidates in the same hierarchy. The improved bidirectional search algorithm was used to simulate this extending process, while music rules were employed to further choose the most appropriate chord sequence. For performance evaluation, a new distance definition was developed to quantify the distance between generated chords and the original chords. The quantitative performance comparisons between the hybrid approach and the traditional HMM approach were also carried out at the end of this paper to show difference between compositional and perceptual process of human beings psychologically.

Study on the context-sensitive Chinese nominal Metaphor Comprehension

Semantic analysis is required in many applications in the field of the natural language processing, such as machine translation, information extraction, etc. Metaphor Comprehension plays an important role in the semantic analysis. In this paper, we use the coincidence degree to find the appropriate meaning of the vehicle in the context. The result shows that the method is efficient.

A computational model of emotion for 3D talking heads

Emotion is an essential element of human-computer interaction, which provides a sympathetic face and human voice for the communication between computers and users. For this purpose, we will describe a computational framework for emotions. Subsequently we will show how this framework can be integrated with an application such as an expressive MPEG-4 3D talking heads with speech synthesis.

Contour detection improved by an easy approach to the perceptional suppression of texture edges

EAAE, the abbreviation for “An easy approach, an acceptable effect”, is becoming popular in current research of computer vision systems. This paper presents a set of perceptional improvements easily implemented in traditional edge and contour detections, such as Cannys approach, by a set of simple methods to a human-like suppression of texture edges. A criterion is also presented for the measure of perceptional capabilities in detection systems kindred.

Cognitive dependency logic

In this article, a novel logic which concerns with the natural property of context is presented, the cognitive state of the agent is considered and the dependency relationships between the word and the sentence, the word and the article, etc. can be inferred. First the cognitive dependency operator is put forward, then the logical system is established, the axioms and the properties of the system is discussed. Finally, the application of the cognitive dependency logic in the understanding of the metaphor is showed.

Anaphoric resolution of Chinese personal pronouns based on the focus-set and DRT

Anaphora is a common phenomenon in a natural language. It plays a large role in the coherence of a text and is a subject of active study in computational linguistics. This paper puts forward several anaphora resolution algorithms for the personal pronouns of written Chinese based on Focus-set Theory and DRT. The strength of our methods lies in the emphasis on the construction of Focus-set and in the dynamic application of syntax knowledge, case grammar and discourse representation theory (DRT). Algorithms were examined by a large Chinese natural language corpus at each stage. The results show that our system of anaphoric resolution algorithms is effective.