Anastasios Venetsanopoulos
Anastasios Venetsanopoulos was a professor of electrical and computer engineering at Toronto Metropolitan University in Toronto, Ontario and a professor emeritus with the Edward S. Rogers Department of Electrical and Computer Engineering at the University of Toronto. In October 2006, Venetsanopoulos joined what was then Ryerson University and served as the founding vice-president of research and innovation. His portfolio included oversight of the university's international activities, research ethics, Office of Research Services, and Office of Innovation and Commercialization. He retired from that position in 2010, but remained a distinguished advisor to the role. Tas Venetsanopoulos continued to actively supervise his research group at the University of Toronto, and was a highly sought-after consultant throughout his career.
Education
Venetsanopoulos received a Bachelor of Electrical and Mechanical Engineering degree from the National Technical University of Athens, and an M.S., M.Phil., and a PhD in electrical engineering from Yale University, New Haven, Connecticut. Tas Venetsanopoulos was a Fulbright Scholar and a Schmitt Scholar, and in 1994 was awarded an honorary doctorate from his alma mater, the National Technical University of Athens.Personal life
In 1986, Venetsanopoulos married Vasiliki Koronakis in a Greek Orthodox service in Toronto, Ontario. They have two daughters: Elizabeth Venetsanopoulos and Dominique Venetsanopoulos.Research interests
Venetsanopoulos' research interests included: biometrics research; multimedia ; digital signal/image processing ; pattern classification and telecommunications.Research record
Venetsanopoulos had a long and productive career in research, education and university administration. He was an internationally renowned researcher in the fields of multimedia systems, digital signal and image processing, digital communications, biometrics and neural networks. Over a period of four decades, he established himself in the worldwide telecommunications and signal processing community as an outstanding researcher, scholar, professor and consultant. He made contributions to telecommunications, signal and image processing, multimedia and biometrics research by authoring and co-authoring many journal papers and books. His pioneering and fundamental research contributions, along with the writing of numerous graduate-level books, opened up new vistas in several fields, including telecommunications; multidimensional filter theory and design; the design of non-linear filters; multimedia neural networks; biometrics applications and WLAN positioning systems.According to a 2020 Google Scholar count, his work has been cited in over 22,600 research papers and 400 textbooks. He was a mentor for over 160 graduate students and post-doctoral fellows.
Telecommunications
Professor Venetsanopoulos' early work dealt with the problem of optimal detection and signal design, to facilitate communication over purely random, general, linear, time-varying, very noisy, undersea acoustic channels. His results contributed to the improvement of SONAR systems for undersea communications over fading dispersive channels and was later applied to ionospheric and tropospheric channels.
Subsequent publications focused on the issue of image and video compression and made contributions in the area of progressive image transmission. PIT refers to the coding of still images at increasing levels of precision. Through PIT, it is possible to expedite activities such as browsing through remote databases of images. Professor Venetsanopoulos developed and tested a number of first and second generation morphological pyramidal techniques, which achieved compression ratios of around 100:1 for good quality, lossy, still image transmission. He contributed to the study of vector quantization for lossy image compression and developed a number of hierarchical coding techniques for still images. Wavelet techniques for still image compression were also addressed by him, as well as fractal-based techniques for compressing and coding still images and video sequences. His later contributions in telecommunications were in the area of mobility management and he developed cost-effective algorithms for mobile terminal location and determination and WLAN positioning systems. This area has attracted interest for its applications in emergency communications, location-sensitive browsing, and resource allocation.
Signal and image processing
Professor Venetsanopoulos was one of the first Canadian researchers to make a contribution to the foundations of two-dimensional and multi-dimensional digital filtering. These techniques are widely used in image and video processing. His early contributions in these areas provided the basis for a variety of techniques that led to efficient two-dimensional filter design. In the eighties, his interest was focused on the area of nonlinear filters. Nonlinear filters are more complex than linear filters but allow additional flexibility and speed in complex applications.
In the area of nonlinear filters, Professor Venetsanopoulos contributed theoretical results, including the introduction of new filter families. The "Nonlinear Order Statistics Filters" were a special case of linear median, order statistics, homomorphic, a-trimmed median, generalized mean, nonlinear mean and fuzzy nonlinear filters. New versions of polynomial filters, such as quadratic filters, were also studied by Professor Venetsanopoulos. He designed new morphological filters, which lead to various detection and recognition applications.
Finally, he conducted extensive research in the area of Adaptive filters. Professor Venetsanopoulos developed Adaptive Order Statistics filters, Adaptive LMS/RLS filters, Adaptive L-filters and Adaptive morphological filter algorithms. These filters are extensively used in numerous biomedical applications, such as in radiology, mammography and tomography. Among other applications, they are also applied to financial data processing and remote sensing.
In the nineties, Professor Venetsanopoulos contributed to the field of color image processing and analysis, where he introduced a number of techniques for color image enhancement filtering and analysis. He also introduced the so-called vector directional filter family, which operates along the direction of the color vectors. A new class of adaptive nonlinear filters was developed. Fuzzy membership functions based on different distance measures were adopted to determine the weights of new nonlinear, adaptive filters. The new filters encompassed different classes of existing nonlinear filters as special cases. For the first time, the color image was treated as a vector field and edge information carried directly by the color vectors was exploited using vector order statistics.
Multimedia signal processing
In 1999 Professor Venetsanopoulos became the Inaugural Chair of the Bell Canada Multimedia Systems Laboratory at the University of Toronto. From that year, he contributed to the area of multimedia data mining and information retrieval by addressing two key technical challenges: a) the problem of similarity determination within the visual data domain, b) the interactive learning of user intentions and automatic adjustment of system parameters for improved retrieval accuracy. He developed still image and video retrieval systems that utilized color content queries. The system implemented a new vector-based approach to image retrieval using an angular-based similarity measure. The scheme he developed addresses the drawbacks of the histogram techniques, it is flexible, and outperforms established retrieval systems. He also developed an interactive learning algorithm for resolving ambiguities arising due to the mismatch between machine-representation of images and human context-dependent interpretation of visual content. His proposed solution exploited feedback from users during retrieval sessions, to adapt their query intentions and improve the accuracy of the retrieved results.
Biometrics research
For thousands of years, humans have used visually-perceived body characteristics such as face and gait to recognize one another. This remarkable ability of human visual system led Professor Venetsanopoulos to build automated systems to recognize individuals from digitally captured facial images and gait sequences. Face and gait recognition belong to the field of biometrics, a very active area of research in computer science, mainly motivated by government and security-related considerations. Face and gait are two typical physiological and behavioral biometrics. Venetsanopoulos contributed to both areas and his research has been extensively cited. There are two general approaches to the subject: the appearance-based approach and the model-based approach.
Appearance-based face recognition processes a 2-D facial image as 2-D holistic patterns. The whole face region is the raw input to a recognition system and each face image is commonly represented by a high-dimensional vector consisting of the pixel intensity values in the image. Thus, face recognition is transformed to a multivariate, statistical pattern recognition problem. In a similar fashion to appearance-based face recognition, an appearance-based gait recognition approach considers gait as a holistic pattern and uses a full-body representation of a human subject as silhouettes or contours. Gait video sequences are naturally three-dimensional objects, formally named tensor objects, and they are very difficult to deal with using traditional vector-based learning algorithms. In order to deal with these tensor objects effectively, Venetsanopoulos and his research team developed a framework of multilinear subspace learning, so that computation and memory demands are reduced, natural structure and correlation in the original data are preserved, and more compact and useful features can be obtained.
The Model-based gait recognition approach considers a human subject as an articulated object, represented by various body poses. Professor Venetsanopoulos proposed a full-body, layered deformable model inspired by the manually labeled body-part-level silhouettes. The LDM has a layered structure to model self-occlusion between body parts and it is deformable, so simple limb deformation is taken into consideration. In addition, it also models shoulder swing. The LDM parameters can be recovered from automatically extracted silhouettes and then used for recognition.