Deep into Deep Learning

The CVML team from CEAS with its latest honor: left to right, Jonathan Song, Associate Professor Siwei Lyu, Assistant Professor Ming-Ching Chang and Yi Wei. Not pictured is Chinese student Lipeng Ke.

ALBANY, N.Y. (December 8, 2017) — For eight years, Computer Vision and Machine Learning (CVML), a joint laboratory of the Computer Science and Electrical Engineering and Computer Science departments, has been the University’s leading research group in artificial intelligence (AI), computer vision and machine learning. It also offers valuable research opportunities for College of Engineering and Applied Sciences (CEAS) students.

This was brought home once again late this summer, when a CVML team that included three students and was led by Assistant Professor Ming-Ching Chang and Associate Professor Siwei Lyu received an honorary mention for overall research contributions in the first IEEE Smart World NVIDIA AI City Challenge in San Jose, California.

Dealing with real-world transportation problems related to safety and congestion, the CVML team’s honors were buttressed by the fact that its model in the Vehicle Detection Challenge was ranked number one in best performance in one test set, and its vehicle detection and tracking results were used as a baseline to support all the other teams in the Smart Transportation Challenge.

UAlbany was among 28 expert teams from 18 universities in San Jose using authentic traffic camera video data and collaborating on projects devised to make transportation systems safer and smarter. Participants tested their algorithms to detect meaningful objects, such as vehicles and pedestrians, and automatically analyze traffic flow based on the detections.

"The technique that the UAlbany team used is a novel method based on deep learning, which solves complicated AI problems with neural networks with many layers of artificial neurons,” said Lyu in alluding to the artificial neural networks that are used in machine learning. Artificial neural networks are akin to biological neural networks such as the human brain, in that they function as a system of interconnected artificial neurons that communicate with each other.

“Deep learning is the same technique behind the recent success of Google AlphaGo and Tesla's self-driving car,” said Lyu, CVML’s director. “Our team has been working at the cutting edge of this exciting technology, and the algorithm we developed is also applicable to many other problems that require automatic and accurate detections — for instance, specific cell types in medical imaging and abnormal patterns in network communication.”

Lyu, of Computer Science, and Chang, of Electrical and Computer Engineering and CVML, co-director, collaborated on the development of their model with computer science doctoral students Yi Wei and Jonathan Song, and Lipeng Ke, a visiting student from the University of Chinese Academy of Sciences in Beijing.

All three students found participation in the competition rewarding. Wei, who learned how to train and tune a deep neural network to detect objects, called it a “great experience” that brought him a true “sense of achievement.”

Song, who trained the object detection model by localizing and identifying multiple objects in a single image, said, “It was amazing experiencing firsthand how creativity and state-of-the-art technology can advance human life by applying artificial intelligence and computer vision on an autonomous car to make our travel safer and more efficient.”

Ke added, “It is a great honor to apply our research to such a competition organized by industry and academia and to be recognized for our achievement in research. I am very proud of our team's achievement.”

IEEE, the Institute of Electrical and Electronics Engineers, is the world’s largest and most prestigious professional organization for electrical and computer engineers and related sciences. NVIDIA, which has funded CVML’s lab generously, is a Portland, Ore.-based technology company that in recent years has delved into deep learning. Chang is a member of IEEE and Lyu an IEEE Senior Member since 2016.

CVML is already signed up to participate next summer in the next round of AI city challenge —“And this time we hope to do better,” Lyu noted.

For more news, subscribe to UAlbany's RSS headline feeds