Quantized Densely Connected U-Nets for Efficient Landmark Localization

ECCV 2018 Zhiqiang TangXi PengShijie GengLingfei WuShaoting ZhangDimitris Metaxas

In this paper, we propose quantized densely connected U-Nets for efficient visual landmark localization. The idea is that features of the same semantic meanings are globally reused across the stacked U-Nets... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Pose Estimation MPII Human Pose DU-Net PCKh-0.5 91.2% # 7