Inyoung Yun

Papers

Total Citations

178

H-Index

About

Inyoung Yun is a leading researcher in efficient deep learning architectures for real-time computer vision, with a primary focus on semantic segmentation for autonomous systems. His most impactful contribution is the development of DABNet (Depth-wise Asymmetric Bottleneck Network), introduced in 2019, which has garnered 178 citations. This work addresses the critical challenge of balancing accuracy and inference speed in pixel-level prediction tasks, proposing a novel lightweight architecture that significantly reduces computational cost and parameters while maintaining high performance—a breakthrough essential for resource-constrained applications like robotics and autonomous driving. Yun’s research has been instrumental in advancing real-time scene understanding, enabling more efficient deployment of deep neural networks on edge devices. By designing specialized bottleneck modules that leverage depth-wise separable convolutions and asymmetric structures, he has helped bridge the gap between model efficiency and predictive power. His work continues to influence the development of fast, accurate vision systems, making him a notable figure in the field of efficient deep learning and embedded computer vision.