Towards Efficient U-Nets: A Coupled and Quantized Approach.

IEEE Transactions on Pattern Analysis and Machine Intelligence(2020)

引用 66|浏览112
暂无评分
摘要
In this paper, we propose to couple stacked U-Nets for efficient visual landmark localization. The key idea is to globally reuse features of the same semantic meanings across the stacked U-Nets. The feature reuse makes each U-Net light-weighted. Specially, we propose an $order$ - $K$ coupling design to trim off long-distance shortcuts, together with an iterative refinement and memory sharing mechanism. To further improve the efficiency, we quantize the parameters, intermediate features, and gradients of the coupled U-Nets to low bit-width numbers. We validate our approach in two tasks: human pose estimation and facial landmark localization. The results show that our approach achieves state-of-the-art localization accuracy but using $\sim 70\%$ fewer parameters, $\sim 30\%$ less inference time, $\sim 98\%$ less model size, and saving $\sim 75\%$ training memory compared with benchmark localizers.
更多
查看译文
关键词
Stacked U-nets,dense connectivity,network quantization,efficient AI,human pose estimation,face alignment
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要