Numerous headlines from the past few years have pointed out challenges attracting, motivating, retaining, and graduating STEM majors, thus leading to worrying shortfalls in these professional areas:
After the tsunami hit Japan on 11 March 2011, in-car navigation systems were the main source of information for those seeking passable roads in the devastated area—impressive testimony to the Japanese IT infrastructure’s extent, degree of integration, and capacity. Tied closely to economic development, success stories in computing research and development in the Asia-Pacific region are plentiful across a diverse array of fields. The June 2012 issue of Computer magazine focused on several of these in its Computing in Asia theme. The featured articles explored a range of topics, from classical research fields such as effective software debugging (from Hong Kong) to cutting-edge areas such as cloud computing (Taiwan), the software development paradigm for internet computing (China), and interdisciplinary bioinformatics genomics research (Singapore).
As computing research and development continue to mature in the Asia-Pacific region, many efforts are driven by the specific needs of local cultures and heritages. Computing Now’s July 2012 theme brings together some articles on this topic from Computer, as well as the increasingly important IEEE International Conference on Asian Language Processing (IALP), featuring research and applications of e-Heritage in China and Cambodia.
In “Computing for the Next-Generation Automobile,” Mikio Aoyama of Japan’s Nanzan University begins with a description of how crowdsourcing using in-car navigation systems assisted in the mapping of passable operational routes after the devastating earthquake and tsunami in 2011. Having established the potential and importance of computing resources in automobiles, the author discusses the state of three important areas of automotive computing: making vehicles greener, making them smarter, and merging transportation and information networks.
In “Computer-Assisted Audiovisual Language Learning,” Lijuan Wang and her colleagues from Microsoft Research Asia discuss the technologies underlying Engkoo, an innovative Web-based computer-assisted audiovisual language-learning service that combines two emerging speech-processing technologies. The system incorporates advanced speech, language, and multimedia technologies to create a virtual tutor that 10 million people in China use to learn English on the Web. Engkoo’s tutoring interaction is modeled on karaoke, a favorite pastime in China, and users learn from a photorealistic lip-synced talking head within a search and discovery ecosystem. The authors describe the text-to-speech architecture, which supports real-time video and audio rendering, and discuss Engkoo’s solution to detecting and correcting input errors due to phonetic mispronunciation.
Initially established as a conference on Chinese computing in 1986, the IALP now focuses primarily on advancing the science and technology of Asian Language Processing. As with the results from Engkoo, however, creative solutions to Asian natural-language processing often prove to be applicable to other more general fields, as well.
In “A Digital Gigapixel Large Format Tile-Scan Camera,” Moshe Ben-Ezra describes a camera assembled at Microsoft Research Asia, using mostly off-the-shelf components, and employed in a cultural-heritage project in Dunhuang cave 46 from the Tang Dynasty. This high-resolution camera is a prototype example of imaging devices that are reasonably simple and cost-effective for digital acquisition and archiving of arts by museums and cultural-heritage sites.
In “Clustering Bayon Face Towers Using Restored 3D Shape Models,” Min Lu and colleagues from Tokyo University, Nara Institute of Science and Technology, and Drexel University describe the adaptation of matrix-recovery theory to recover incomplete archaeological data. They used hierarchical clustering of similar 3D scanned data to analyze the structure and patterns of face towers at the Khmer temple at Angkor, Cambodia.
Zhigeng Pan and colleagues’ “Animating and Interacting with Ancient Chinese Painting — Qingming Festival by the Riverside,” describes a novel approach to extracting, reconstructing, and animating 3D characters to allow viewers to interact with the characters from an ancient painting.
In all these cases, the research and development efforts are driven by local cultural and heritage needs. Though the end results themselves are significant, the scientific discoveries and contributions along the way are more broadly applicable to the rest of the world and are thus the real contributions.
Kelvin Sung is a professor of computing and software systems at University of Washington, Bothell. He has a PhD in computer science from the University of Illinois, Urbana-Champaign. His interests include computer graphics, all aspects of image synthesis, and serious games development. Contact him at firstname.lastname@example.org.