Research

Mon, 01 Jan 0001 00:00:00 +0000

My work studies the intelligence underlying human behavior, performance, and interaction, with generative models as the primary instrument: from monocular motion capture and real-time character control to today’s large-scale, multimodal generative models. The projects and publications below trace this line of work over the years.

I also serve as a reviewer for conferences and journals including: SIGGRAPH, SIGGRAPH Asia, Eurographics, Pacific Graphics, NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, AAAI; TOG, TVCG, CGF, T-PAMI, TIP, and others.

Mendocino: The End of Highway 1

Sun, 06 Sep 2026 00:00:00 +0000

I first heard the word Mendocino while I was learning about electromagnetic induction. Somewhere online I came across a small device called a Mendocino motor: sunlight falls on the solar cells around its four sides, the coils energise in turn, and the rotor, floating in a magnetic field, spins with almost no friction. It was the clearest picture I had when I finally worked out that free energy is not the same thing as perpetual motion.

Home on Mingyi Shi (史明镒)

Research

Mendocino: The End of Highway 1