Papers
2
Total Citations
23
H-Index
2
About
Ying Xie is a rising researcher at the intersection of brain–computer interfaces (BCIs) and computer vision, whose work bridges neural signal processing and real-time visual understanding. In their most cited paper (2021, 21 citations), Xie tackled one of BCI’s hardest problems—cross-session and cross-subject variability in motor imagery EEG classification—by introducing a transfer learning method based on rotation alignment with Riemannian mean. This approach significantly improved the robustness of wheelchair control systems, demonstrating how Riemannian geometry can stabilize neural decoding across different users and recording sessions. More recently, Xie’s 2025 work on FRISNET (2 citations) pushes the boundaries of real-time instance segmentation by fusing frequency-domain and multilevel spatial features, addressing the challenge of accurate mask prediction in complex scenes. This dual expertise—from decoding brain signals to parsing visual scenes—positions Xie as a versatile contributor to assistive robotics and embodied AI. Their research exemplifies how geometric deep learning and frequency analysis can solve domain-shift problems in both neural and visual domains, with clear implications for autonomous systems that must adapt to new users and environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2