To my son to remind me of my dreams; to my husband to support me in pursuing my dreams; to my mother to guide me towards my dreams; to my family and friends to tell me to believe in my dreams; to my colleagues to help me realize my dreams on professional level.

SccF-F low bit-rate video cdeo- is vital for the transmission of video signals over wireless crelessAscjNjx - model-based videoceo-w sco-w is proposed in this paper toachVFV this. This paper mainly addressesautomatic sctomat fac model design. Firstly, a robust and adaptivefac segmentation method is proposed,whic is based on piecVVFE skin-cE-JT distributions. 43 million skin pixels from 900 images are used to train theskin-cE-JT model, whic ci identifyskin-cwwxx pixels reliably under different lightingchting-TFN Next, reliable algorithms are proposed fordetechTthe eyes, mouth and cd- that are used to verify the fac chVTEE-JhVF Then, based on thedetecwj facc features and human fac muscwhw distributions, aheuristic scristi fac model is designed to represent the rigid and non-rigid motion of head andfacwj features. A novel motion estimation algorithm is proposed to estimate theobjec model motion hierarcxx-JhNE Experimental results are provided to illustrate theperformanc of the proposed algorithms forfacFw feature detece-w and theac-hhVV of the designed scigned fac model for representingfac motion.

We present in this paper a CBIR system for use in a psychological study of the relationship between human movement and Dyslexia. The system allows access to up to 500 hours of video and is an example of a scientific user context. This user context requires 100% accurate indexing and retrieval for a set of specific queries. This paper presents a novel use of interactive visual and audio cues for attaining this level of indexing performance.

Abstract—This work presents a low-complexity system implementation of a novel content-adaptive background skipping scheme for region-of-interest (ROI) video coding in mobile video phone applications. To improve the overall perceptual quality, the proposed approach reallocates bits to ROI macroblocks by adaptively skipping the non-ROI and using a weighted bit allocation scheme. The skip decision for the non-ROI is adaptively determined by the content information of the video, such as foreground shape deformation, foreground motion, background motion and background texture complexity. Experimental results demonstrate that the proposed scheme outperforms traditional schemes by up to 2 dB.

In this paper, we propose a novel region-of-interest (ROI) video coding algorithm for wireless video telephony applications. In order to improve the visual quality of the ROI, the proposed approach reallocates bits from Non-ROI macroblocks to ROI by adaptively skipping Non-ROI and using an optimized weighted bit allocation scheme to bias the bit allocation. To the best of our knowledge, this work is the first effort to develop an optimized ρ-domain bit allocation scheme for ROI video coding. Experimental results indicate that the proposed approach significantly outperforms other methods by up to 2dB. 1.

Abstract: This paper presents a user interactive system that manages, with minimum user interaction, to locate a set of characteristic feature points in the frontal view of a person’s face. The proposed method is user adaptive and is implemented as a combination of a classification-based facial region extraction technique with a transition based facial feature extraction method. Thus, a feed-forward neural network is first trained to recognize image blocks belonging to the facial area. Then, the feature detection algorithm is initialized with a blink of the user’s eyes, which localizes the exact contours of the eyes. Finally, we locate the remaining facial features (face contour, eyebrows, mouth and nose) by adaptively thresholding the facial area and utilizing apriori knowledge about the geometrical structure of the face.

The amount of digital video has been increasing dramatically due to the technology advances in video capturing, storage, and compression. The usefulness of vast repositories of digital information is limited by the e#ectiveness of the access methods, as shown by the Web explosion. The key issues in addressing the access methods are those of content description and of information space navigation. While textual documents in digital form are somewhat self-describing (i.e., they provide explicit indices, such as words and sentences that can be directly used to categorise and access them), digital video does not provide such an explicit content description. In order to access video material in an e#ective way, without looking at the material in its entirety, it is therefore necessary to analyse and annotate video sequences, and provide an explicit content description targeted to the user needs.

Nowadays most web pages contain both text and images. Nevertheless, search engines index documents based on their disseminated content or their meta-tags only. Although many search engines offer image search, this service is based over textual information filtering and retrieval. Thus, in order to facilitate effective search for images on the web, text analysis and image processing must work in complement. This paper presents an enhanced information fusion version of the meta-search engine proposed in [1], which utilizes up to 9 known search engines simultaneously for content information retrieval while 3 of them can be used for image processing in parallel. In particular this proposed meta-search engine is combined with fuzzy logic rules and a neural network in order to provide an additional search service for human photos in the web.

The field of automated video surveillance has experienced increased research interest due to falling costs of video sensors, increasing security concerns, and the need for improved algorithm for extracting high-level information from video sequences. The analysis of human activities and their environment within the context of security provides information enabling the proactive identification of anomalous behavior. This makes human detection a prerequisite for the automatic extraction of higher level information, such as the recognition of the activities of individual humans. In this paper, we approach the challenge of detecting humans within video sequences as a classification task; moving objects in the foreground are either human or non-human. The classification approach presented in this work is based on motion (periodic motion detection), appearance (skin color detection), and shape (MPEG-7 shape descriptors). A modular infrastructure for data collection, object instantiation, and tracking was also implemented.

Our experiments in TRECVID 2008 include participation in the high-level feature extraction, automatic search, video summarization, and video copy detection tasks, using a common system framework. In the high-level feature extraction task, we extended our last year’s experiments, which were based on SOM-based semantic concept modeling followed by a post-processing stage utilizing the concepts ’ temporal and inter-concept co-occurrences. We also studied the effects of a more comprehensive feature selection and the inclusion of audio features and face detection. We submitted the following six runs: • A_PicSOM_1_6: Visual features, baseline feature selection • A_PicSOM_2_2: Visual features, baseline feature selection, temporal context