V. LEARNING, RECOGNITION AND HUMAN COMPUTER INTERACTION

The objective of this work is to find and use relationships between low level image structure and higher order representations and recognition, and using these relationships as aids in human computer interaction.

V.1. We explore new algorithms for computer vision based on multilinear algebra. Firstly, we learn the expression subspace and person subspace from a corpus of images based on Higher-Order Singular Value Decomposition (HOSVD), and investigate their applications in facial expression synthesis, face recognition and facial expression recognition. Secondly, we explore new algorithms for image ensembles/video representation and recognition using tensor rank-one decomposition and tensor rank-R approximation.

V.2. Dynamic shape model and its application in fire detection

V.3. Recognition is achieved either by explicitly coding the recognition criteria in terms of low level structure, or through learning from examples. Learning algorithms incorporate subspace projections of higher dimensional data symbolically or using neural approaches.

V.4. The aforementioned work on representation and learning has contributed to two types of human computer interfaces we have developed. First, learning and classification techniques, including usual statistical classifiers, neural networks, support vector machines and artificial intelligence approaches, have been used to develop new methods for human face detection and hand gesture recognition.

V.5. The second type of human-computer interface is a free-hand-sketch based interface for image editing (e.g., moving, size-scaling, color-transforming parts of an image) is developed. The sketches drawn by the user on top of the image serve as a natural way of specifying an image part and the editing (e.g., move, deletion) operation to be performed.

V.6. Low level image features are used for the problem of object categorization. In general, object categorization comprises two main research areas:  (1) classification or clustering of images containing objects  belonging to an object category, and (2) detection, localization, and  segmentation of individual object-category instances in images. The  first thrust of research is typically concerned with exemplar based  methods, where the main focus is to develop an efficient distance  measure between two images. Work in the second research area is  primarily concerned with object-category modeling on training images,  and using category models for object detection, localization and  segmentation in test images. These approaches differ from object  recognition methods in that category instances in the training and  test sets are different.