Fanman Meng

University of Electronic Science and Technology of China

Papers

Total Citations

H-Index

About

Fanman Meng is a leading researcher in computer vision and multimodal perception, with a focus on robust visual understanding under challenging real-world conditions. His work spans pointer meter reading, audio-visual segmentation, and cross-modal learning. Meng’s most cited paper, "Reading Various Types of Pointer Meters Under Extreme Motion Blur" (2023, 17 citations), addresses a critical gap in automated inspection systems—persistent motion blur from moving platforms like drones and patrol robots. By developing deep learning methods that maintain high precision despite camera shake, he has advanced industrial automation and robotics. More recently, his 2024 paper "Cross-Modal Cognitive Consensus Guided Audio–Visual Segmentation" (4 citations) pioneers a cognitive consensus framework for audio-visual segmentation (AVS), enabling pixel-level extraction of sounding objects from video frames. This work has immediate applications in multi-modal video editing, augmented reality, and intelligent robotics. Meng’s contributions are notable for bridging theoretical innovation with practical deployment, and his research continues to shape how machines perceive and interact with dynamic, noisy environments—making him a key figure in the evolution of robust, multimodal AI systems.