Papers
4
Total Citations
63
H-Index
3
About
Sergey Kulik is a researcher at the forefront of intelligent robotics and computer vision, with a particular focus on making neural network systems practical for real-world robotic applications. His most impactful work centers on the deployment of convolutional neural networks (CNNs) and object detection systems, such as YOLO, under challenging conditions—specifically when training data is scarce. His 2020 paper on training YOLO with small datasets (28 citations) directly addresses a critical bottleneck in robotics: the need for robust perception without massive, labeled image collections. Expanding on this, his highly cited work on using CNNs for recognizing objects with highly varied appearances (26 citations) tackles the fundamental problem of visual generalization, enabling robots to reliably identify objects that look different from their training examples. Kulik has also ventured into affective computing, exploring emotion detection in illustrations via CNNs, and has contributed to optimization in robotics through genetic algorithms. His research is distinguished by its pragmatic, problem-driven approach—solving the real-world constraints of data, variability, and computational efficiency that limit the deployment of intelligent robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4