Minyi Guo
Papers
6
Total Citations
54
H-Index
3
About
Minyi Guo is a prominent researcher whose work sits at the intersection of computer vision, autonomous systems, and mobile computing. His research has made significant contributions to monocular 3D object detection (Mono3D), a field concerned with accurately identifying and localizing objects in three-dimensional space using a single camera — a critical capability for vehicles, drones, and robots operating under real-world computational constraints. Among his most impactful contributions is the MonoATT framework, which introduced an Adaptive Token Transformer to address the limitations of grid-based vision tokens in resource-constrained mobile environments, accumulating 37 citations and establishing a new direction for online Mono3D systems. He has further advanced the field through ground depth estimation techniques — explored in both MoGDE and subsequent work — which tackle the persistent near-far disparity challenge inherent to monocular vision. Beyond perception, Guo has explored trajectory classification for mobile IoT devices and accelerator-level parallelism for autonomous micromobility systems, reflecting a broad systems-oriented perspective on edge AI deployment. His body of work demonstrates a consistent commitment to bridging the gap between high-accuracy deep learning models and the practical demands of real-world autonomous platforms, making his research highly relevant for engineers and scientists working on next-generation intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6