LCSCNet: Linear Compressing Based Skip-Connecting Network for Image Super-Resolution

9 Sep 2019  ·  Wenming Yang, Xuechen Zhang, Yapeng Tian, Wei Wang, Jing-Hao Xue, Qingmin Liao ·

In this paper, we develop a concise but efficient network architecture called linear compressing based skip-connecting network (LCSCNet) for image super-resolution. Compared with two representative network architectures with skip connections, ResNet and DenseNet, a linear compressing layer is designed in LCSCNet for skip connection, which connects former feature maps and distinguishes them from newly-explored feature maps. In this way, the proposed LCSCNet enjoys the merits of the distinguish feature treatment of DenseNet and the parameter-economic form of ResNet. Moreover, to better exploit hierarchical information from both low and high levels of various receptive fields in deep models, inspired by gate units in LSTM, we also propose an adaptive element-wise fusion strategy with multi-supervised training. Experimental results in comparison with state-of-the-art algorithms validate the effectiveness of LCSCNet.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Super-Resolution BSD100 - 3x upscaling LCSCNet PSNR 28.87 # 15
Image Super-Resolution Set14 - 3x upscaling LCSCNet PSNR 29.87 # 14
Image Super-Resolution Set5 - 3x upscaling LCSCNet PSNR 33.99 # 18
Image Super-Resolution Urban100 - 3x upscaling LCSCNet PSNR 27.24 # 15

Methods