Papers

17

Total Citations

912

H-Index

12

About

Martin D. Levine is a pioneering computer vision and robotics researcher whose work spans image segmentation, biologically inspired vision systems, and intelligent robotic applications. Based at McGill University, Levine made an early and lasting impact with his 1984 expert system approach to low-level image segmentation, which applied rule-based reasoning to the complex problem of parsing natural scenes — a contribution that has garnered over 320 citations and remains a foundational reference in the field. His 1973 work on computer-determined depth maps further established his credentials as an early innovator in machine perception. A recurring theme throughout Levine's career is the translation of biological vision principles into practical robotic systems. His research on foveated imaging — sensors that mimic the primate visual system's non-uniform spatial resolution — produced influential reviews, hardware implementations in standard CMOS technology, and real-time attentional algorithms. These contributions, reflected across multiple cited works from the mid-1990s through early 2000s, helped define the subfield of space-variant robotic vision. Levine also applied his expertise to industrial automation, developing intelligent arc welding robots capable of joint recognition and adaptive process control. Collectively, his body of work bridges artificial intelligence, neuroscience-inspired sensing, and practical robotics engineering.

Research Focus

Key Achievements

12
H-Index
17
Papers
912
Total Citations
54
Avg Citations/Paper
🏆 Most Cited Paper
Low Level Image Segmentation: An Expert System
320 citations · 1984
📈 Most Prolific Year: 2002 (4 Papers)
🤝 Key Collaborators: 14
🏛 Institutions: McGill University, Jet Propulsion Laboratory, California Institute of Technology

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago