Haifeng Gong

University of Pennsylvania, Google (United States)

Papers

Total Citations

108

H-Index

About

Haifeng Gong is a computer vision researcher whose work centers on motion estimation and visual object tracking, particularly in challenging, real-world environments. His most significant contribution, the 2011 paper "Multi-hypothesis motion planning for visual object tracking," has garnered 105 citations and addresses a critical problem in the field: persistent occlusions in crowded street scenes. Drawing inspiration from robot motion planning, Gong proposed a long-term motion model that maintains multiple trajectory hypotheses, enabling robust tracking even when targets are temporarily hidden. This innovative approach bridges the gap between robotics and computer vision, offering a principled solution to a long-standing tracking challenge. Gong's work is notable for its practical impact on surveillance, autonomous driving, and human-computer interaction, where reliable object tracking is essential. While his later research on real-time one-dimensional motion estimation (2015) has seen limited citations, it reflects his ongoing interest in efficient, low-latency algorithms for embedded vision systems. Gong's contributions exemplify how cross-disciplinary thinking can advance core computer vision tasks, making his research valuable for students and engineers developing robust tracking systems for dynamic, unpredictable scenes.