Songsong Xiong
Papers
3
Total Citations
19
H-Index
2
About
Songsong Xiong is an emerging researcher specializing in robotic perception, computer vision, and machine learning, with a particular focus on advancing object recognition capabilities for service robots operating in real-world human-centered environments. His work addresses one of the most persistent challenges in robotics: enabling machines to accurately distinguish between visually similar objects across diverse and dynamic contexts. Xiong's most notable contribution lies in developing hybrid multi-modal architectures that combine Vision Transformers and Convolutional Neural Networks for fine-grained 3D object recognition — work that has garnered 12 citations across two related publications and demonstrates his commitment to iterative, rigorous research. His investigations into lifelong ensemble learning represent another significant thread of his scholarship, proposing novel frameworks that allow robots to continuously acquire knowledge about new objects in an open-ended fashion while retaining previously learned representations — a critical capability for practical deployment of service robots. With a research agenda that bridges deep learning methodology and real-world robotic applications, Xiong is positioning himself as a thoughtful contributor to the field of few-shot and continual learning. His work holds meaningful implications for robotics in retail, hospitality, and domestic settings, where adaptability and precision are paramount.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3