Herui Heng
Papers
2
Total Citations
14
H-Index
2
About
Herui Heng is a researcher focused on advancing scene text recognition, particularly for challenging, real-world applications in the logistics industry. Their core contributions lie in developing robust models that can accurately interpret text from low-quality, curved, or distorted images—a critical need for automating express sheet processing. Heng’s most cited work, a 2022 study on context modeling for low-resolution logistics images, has garnered 12 citations and demonstrates a practical approach to overcoming visual degradation. Building on this, their 2023 paper on MTSTR introduces a multi-task learning framework enhanced by a dual attention mechanism, pushing the boundaries of performance in complex, low-resolution scenarios. By targeting the intersection of computer vision and industrial automation, Heng’s research directly addresses the gap between academic text recognition and the messy realities of logistics environments. Their work not only advances technical methodologies but also provides tangible solutions for improving efficiency in supply chain operations, making them a notable contributor to applied AI in industry.
Research Focus
Key Achievements
Top Papers
- 1
- 2