Zhu Songchun

Overview

Zhu Songchun is a computer vision researcher whose career has run through universities in the United States and, later, research institutions in China. He joined the University of California, Los Angeles in 2002 and worked there as a computer vision researcher, and in September 2020 Peking University announced that he would lead its Institute for Artificial Intelligence.^[1] He was also involved in setting up the Beijing Institute for General Artificial Intelligence (BIGAI).^[1]

Computer vision, the field in which most of his listed work sits, is the part of artificial intelligence concerned with getting machines to extract information from images and video: recognising objects, recovering the layout of a scene, and turning pixels into descriptions a program can act on. It borrows from geometry, statistics and machine learning, and it is one of the areas where AI research most often meets cameras, medical imaging and robotics.

Education and early research

Zhu studied as an undergraduate at the University of Science and Technology of China, and received his PhD from Harvard University in 1996.^[1] After the doctorate he held brief posts at Stanford University and Ohio State University.^[1] Those early appointments are recorded without dates, but they place him for a period in the American academic system before he moved to California.

At UCLA and the Marr Prize

In 2002 he joined the University of California, Los Angeles, where he worked as a computer vision researcher.^[1] One year after joining UCLA he received the Marr Prize, which the profile describes as the highest honour in computer vision.^[1] Prizes of this kind are how a research community signals which pieces of work it considers to have moved the field; for a researcher, receiving one early in a new post marks recognition from peers working on the same problems.

Institutions in China

In September 2020, Peking University announced that Zhu would lead its Institute for Artificial Intelligence.^[1] The announcement put him at the head of a university-level institute rather than a single laboratory, a role that involves setting research directions and building a group rather than only carrying out his own projects.

He was also involved in setting up the Beijing Institute for General Artificial Intelligence, known as BIGAI.^[1] The name points at a specific research goal: general artificial intelligence is usually contrasted with narrow systems that are built for one task, and describes the aim of systems that can handle a wide range of tasks and adapt to situations they were not specifically designed for. An institute organised around that goal is a bet that progress needs sustained, shared infrastructure rather than isolated projects.

Significance

The record of Zhu's career is a sequence of stages: an undergraduate degree at the University of Science and Technology of China, a Harvard PhD in 1996, short appointments at Stanford and Ohio State, a long stretch as a computer vision researcher at UCLA from 2002, and then, from 2020, leadership of Chinese AI institutions.^[1] The Marr Prize sits at the point where the research career and the institutional one overlap, and his later work has been as much about building organisations as about individual results.^[1]