Seiji Ishikawa

Kyushu Institute of Technology, Juntendo University

Papers

10

Total Citations

77

H-Index

5

About

Seiji Ishikawa is a computer vision and robotics researcher whose work spans human motion recognition, autonomous robot systems, and multi-robot cooperation. His most significant contributions lie in the development of efficient techniques for recognizing and segmenting human activities, particularly through optical flow-based representations such as directional motion history images and energy images. His 2009 and 2010 papers on temporal motion recognition and combined directional motion cues — garnering 16 and 15 citations respectively — established practical frameworks for decomposing complex activities into recognizable action primitives, advancing the field of appearance-based action understanding. Ishikawa also made early contributions to eigenspace-based real-time motion recognition, demonstrating its viability in surveillance contexts including aerial robot platforms. His robotics research spans decades, from foundational 1997 work on cooperative multi-robot architectures to practical implementations of two-robot cooperative conveyance and face-tracking autonomous systems. More recently, his human care robot work demonstrates a commitment to socially beneficial applications of intelligent robotics. Though his citation counts reflect a specialized academic audience, Ishikawa's body of work represents a sustained, multidisciplinary contribution to computer vision, human-robot interaction, and autonomous systems research over more than two decades.

Research Focus

Key Achievements

5
H-Index
10
Papers
77
Total Citations
8
Avg Citations/Paper
🏆 Most Cited Paper
Temporal motion recognition and segmentation approach
16 citations · 2009
📈 Most Prolific Year: 2009 (1 Papers)
🤝 Key Collaborators: 24
🏛 Institutions: Kyushu Institute of Technology, Juntendo University

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 · 15 days ago