Weihua Ou

Papers

Total Citations

H-Index

About

Weihua Ou is a leading researcher in lightweight deep learning and multimedia signal processing, with a focus on semantic segmentation and model compression. His most cited work, "Boundary-Guided Lightweight Semantic Segmentation With Multi-Scale Semantic Context" (2024, 65 citations), introduces a novel dual-resolution architecture that efficiently encodes both fine-grained image details and high-level semantics. This method is critical for real-time applications such as autonomous driving, robotic vision, and virtual reality, where computational efficiency and accuracy must coexist. Ou’s contributions address a key bottleneck in deploying deep neural networks on resource-constrained devices. In his earlier work, "Feature fusion-based collaborative learning for knowledge distillation" (2021, 2 citations), he explored advanced model compression techniques, proposing a collaborative distillation framework that enhances the performance of compact student models by fusing multi-level features from a teacher network. This research is foundational for training efficient deep models without sacrificing accuracy. Ou’s work has been recognized for its practical impact on intelligent systems, bridging the gap between cutting-edge deep learning theory and real-world multimedia applications. His research continues to influence the development of lightweight, high-performance neural architectures for edge computing.