PhD Student in Computer Vision
Communication Engineering Background | Full-Stack Algorithm Enthusiast
Currently hacking: Computer VisionDeep LearningImage Processing

Tech Stack: Python • MMyolo • PyTorch
Features:

• Self-attention + Convolution hybrid architecture
• Real-time object detection (30+ FPS)
• Advanced occlusion handling

Citation

@article{guo2025vehicle,
title={Vehicle detection in CCTV with global-guided self-attention and convolution},
author={Guo, Yupei and Yamamoto, Yota and Yaginuma, Hideki and Taniguchi, Yukinobu},
journal={Complex \& Intelligent Systems},
volume={11},
number={10},
pages={458},
year={2025},
publisher={Springer}
}

Tech Stack: Python • Ultralytics • PyTorch
Features:

• Spatial and Frequency downsampling design
• SOTA precsion results on both CCTV and UAV
• Competitive latency for real-time detection

Citation

@inproceedings{guo2026separation,
title={Separation-Aware Downsampling for Vehicle Detection.},
author={Guo, Yupei and Yamamoto, Yota and Taniguchi, Yukinobu},
booktitle={VISAPP (1)},
pages={483--490},
year={2026}
}

PhD Computer science
Research Focus:

• Multi-agent
• Vision Language model

MSc Information and Computer Technology
Research Focus:

• Object Detection
• Generative AI

BEng Communication Engineering
Thesis Work:

• Signal Processing Algorithms
• Wireless Communication Systems