Mr. Tianhong Qi | Meteorological AI Applications | Innovative Research AwardΒ
Student atΒ University of Science and Technology Beijing , China.
Tianhong Qi is a talented and driven undergraduate student at the School of Intelligence Science and Technology, University of Science and Technology Beijing π. With a passion for scientific research, Tianhong has made significant contributions to various fields, including intelligent image recognition, object detection, and data mining π€.
Professional Profile
π Education
– *Primary School*: Graduated from Beijing Zhongguancun No.3 Primary School π– *Middle School*: Graduated from the Humanities Experimental Class of Beijing 101 High School π– *University*: Currently an undergraduate student at the School of Intelligence Science and Technology, University of Science and Technology Beijing π
πΌ Experience
– *Chinese Cursive Script Recognition*: Built a cursive script dataset and designed a skeleton-embedded Vision Transformer (ViT) model with progressive curriculum learning π– *Ground-Based Cloud Detection*: Proposed CurriCloud, a novel model integrating meteorological theory and loss-adaptive curriculum learning βοΈ– *Intelligent Archival Document System*: Developed a multi-modal system for document recognition, automatic classification, and knowledge graph construction π– *Digital Twin for Smart Manufacturing*: First-authored paper on digital twin frameworks π€– *ZhiYuan Interactive Platform*: Created a multi-functional app integrating DFS-based puzzle games, real-time chat, and file transfer π±– *AI Marathon Training System*: Combined biomechanics, meteorology, and graph computing to optimize training plans πββοΈ
π¬ Research Interests
-Tianhong’s research interests lie in applying computer vision and data mining to scientific domains, including:
π Awards
– Winner of multiple municipal and national youth and university-level science and technology innovation competitions π
– Winner of the 2025 “Challenge Cup” Capital Academic Competition π
– Prize winner of the 2024 “Cradle Cup” Innovation Competition (USTB)
πTop NotedΒ Publications
– Vision Transformer-Based Recognition of Chinese Cursive Calligraphy: A Curriculum Learning and Skeleton Embedding Approach π
– Loss Adaptive Curriculum Learning for Ground-Based Cloud Detection βοΈ
– Digital Twin Frameworks for Smart Manufacturing π€
– Judicial Evidence Relationship Construction Method and Device (National patent) π
– Intelligent Archival System V1.0 (Software copyright) π
– ZhiYuan Interactive Platform (Software copyright) π±