Yuntian Feng

Papers

Total Citations

H-Index

About

Yuntian Feng is a researcher in mobile robotics and computer vision, with a focus on visual place recognition and efficient neural network architectures. His most-cited work, "Salient Feature Selection for CNN-Based Visual Place Recognition" (2018), addresses a critical challenge in deploying convolutional neural networks for real-time robotic navigation: the high dimensionality of CNN-based image representations. Feng proposed a salient feature selection method that reduces computational overhead while maintaining recognition accuracy, enabling faster and more reliable place recognition in large-scale, dynamic environments. This contribution is particularly valuable for autonomous robots operating in real-world settings, where speed and robustness are paramount. Though his citation count is still growing, Feng's work has been recognized for its practical impact on mobile robot localization and mapping. His research bridges the gap between deep learning performance and real-time system constraints, offering a pathway for more efficient visual perception in robotics. As the field moves toward lightweight, deployable AI, Feng's contributions to feature selection and representation learning continue to inform new approaches in visual place recognition.