Nicolai Petkov

University of Groningen

Papers

17

Total Citations

396

H-Index

12

About

Nicolai Petkov is a researcher whose work spans computer vision, robotics, and biologically inspired computational models, with particular emphasis on visual place recognition, semantic mapping, and object detection. His most influential contributions center on applying convolutional neural networks to appearance-invariant place recognition — a critical challenge in autonomous driving and mobile robotics — with his 2017 paper accumulating 78 citations and a related 2015 study earning 52, reflecting sustained community interest in this direction. Petkov has also made notable strides in modeling the primate visual system, proposing the trainable S-COSFIRE hierarchical object recognition framework and extending COSFIRE-based filters to crack delineation and curved line detection. His robotics research is notably interdisciplinary, encompassing semantic indoor mapping through the ViMantic architecture, human action recognition from skeletal data, and even odor recognition for robotic applications. His involvement in the TrimBot2020 project demonstrates a commitment to real-world deployment of intelligent outdoor robots. Collectively, his portfolio reflects a researcher who bridges neuroscience-inspired modeling with practical autonomous systems, earning over 300 cumulative citations and contributing meaningfully to both foundational theory and applied robotics.

Research Focus

Key Achievements

12
H-Index
17
Papers
396
Total Citations
23
Avg Citations/Paper
🏆 Most Cited Paper
Appearance-invariant place recognition by discriminatively training a convolutional neural network
78 citations · 2017
📈 Most Prolific Year: 2017 (3 Papers)
🤝 Key Collaborators: 31
🏛 Institutions: University of Groningen

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