Timezone: »

Finding Discriminative Filters for Specific Degradations in Blind Super-Resolution
Liangbin Xie · Xintao Wang · Chao Dong · Zhongang Qi · Ying Shan

Wed Dec 08 12:30 AM -- 02:00 AM (PST) @

Recent blind super-resolution (SR) methods typically consist of two branches, one for degradation prediction and the other for conditional restoration. However, our experiments show that a one-branch network can achieve comparable performance to the two-branch scheme. Then we wonder: how can one-branch networks automatically learn to distinguish degradations? To find the answer, we propose a new diagnostic tool -- Filter Attribution method based on Integral Gradient (FAIG). Unlike previous integral gradient methods, our FAIG aims at finding the most discriminative filters instead of input pixels/features for degradation removal in blind SR networks. With the discovered filters, we further develop a simple yet effective method to predict the degradation of an input image. Based on FAIG, we show that, in one-branch blind SR networks, 1) we could find a very small number of (1%) discriminative filters for each specific degradation; 2) The weights, locations and connections of the discovered filters are all important to determine the specific network function. 3) The task of degradation prediction can be implicitly realized by these discriminative filters without explicit supervised learning. Our findings can not only help us better understand network behaviors inside one-branch blind SR networks, but also provide guidance on designing more efficient architectures and diagnosing networks for blind SR.

Author Information

Liangbin Xie (Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Chinese Academy of Sciences)
Xintao Wang (Applied Research Center, Tencent PCG)
Chao Dong (Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Chinese Academy of Sciences)
Zhongang Qi (Oregon State University)
Ying Shan (Tencent)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors

  • 2020 Poster: Detecting Interactions from Neural Networks via Topological Analysis »
    Zirui Liu · Qingquan Song · Kaixiong Zhou · Ting-Hsiang Wang · Ying Shan · Xia Hu
  • 2017 : Posters 1 »
    J.P. Lewis · Housam Khalifa Bashier Babiker · Zhongang Qi · Laura Rieger · Ning Xie · Filip Dabek · Koushik Nagasubramanian · Bolei Zhou · Dieuwke Hupkes · CHUN-HAO CHANG · Pamela K Douglas · Enea Ceolini · Derek Doran · Yan Liu · Fuxin Li · Randolph Goebel