000262784 001__ 262784
000262784 005__ 20190619220149.0
000262784 037__ $$aCONF
000262784 245__ $$aA comparative study on wavelets and residuals in deep super resolution
000262784 260__ $$c2019
000262784 269__ $$a2019
000262784 300__ $$a6
000262784 336__ $$aConference Papers
000262784 520__ $$aDespite the advances in single-image super resolution using deep convolutional networks, the main problem remains unsolved: recovering fine texture details. Recent works in super resolution aim at modifying the training of neural networks to enable the recovery of these details. Among the different method proposed, wavelet decomposition are used as inputs to super resolution networks to provide structural information about the image. Residual connections may also link different network layers to help propagate high frequencies. We review and compare the usage of wavelets and residuals in training super resolution neural networks. We show that residual connections are key in improving the performance of deep super resolution networks. We also show that there is no statistically significant performance difference between spatial and wavelet inputs. Finally, we propose a new super resolution architecture that saves memory costs while still using residual connections, and performing comparably to the current state of the art.
000262784 6531_ $$asuper resolution, deep learning, wavelet decomposition, residual learning
000262784 700__ $$g242277$$aZhou, Ruofan$$0249528
000262784 700__ $$g254809$$aLahoud, Fayez$$0254717
000262784 700__ $$aEl Helou, Majed$$0250358$$g266645
000262784 700__ $$0241946$$aSüsstrunk, Sabine$$g125681
000262784 7112_ $$a2019 IS&T International Symposium on Electronic Imaging$$cBurlingame, California USA$$d13 - 17 January, 2019
000262784 773__ $$tIS&T EI Proceedings
000262784 8560_ $$ffayez.lahoud@epfl.ch
000262784 8564_ $$uhttps://infoscience.epfl.ch/record/262784/files/manuscript.pdf$$zFinal$$s3158892
000262784 909C0 $$xU10429$$pIVRL$$msabine.susstrunk@epfl.ch$$zGrolimund, Raphael$$0252320
000262784 909CO $$qGLOBAL_SET$$pconf$$pIC$$ooai:infoscience.epfl.ch:262784
000262784 960__ $$afayez.lahoud@epfl.ch
000262784 961__ $$afantin.reichler@epfl.ch
000262784 973__ $$rREVIEWED$$aEPFL
000262784 981__ $$aoverwrite
000262784 980__ $$aCONF