Can I read Multi-Temporal Recurrent Neural Networks For Progressive Non-Uniform Single Image Deblurring With Incremental Temporal Training on EtoBox?

Multi-Temporal Recurrent Neural Networks For Progressive Non-Uniform Single Image Deblurring With Incremental Temporal Training by Park, Dongwon; Kang, Dong Un; Kim, Jisoo; Chun, Se Young is a scholarly article available to read on EtoBox.

What is Multi-Temporal Recurrent Neural Networks For Progressive Non-Uniform Single Image Deblurring With Incremental Temporal Training about?

Multi-scale (MS) approaches have been widely investigated for blind single image / video deblurring that sequentially recovers deblurred images in low spatial scale first and then in high spatial scale later with the output of lower scales. MS approaches have been effective especially for severe blurs induced by large motions in high spatial scale since those can be seen as small blurs in low spatial scale. In this work, we investigate alternative approach to MS, called multi-temporal (MT) approach, for non-uniform single image deblurring. We propose incremental temporal training with constructed MT level dataset from time-resolved dataset, develop novel MT-RNNs with recurrent feature maps, and investigate progressive single image deblurring over iterations. Our proposed MT methods outperform state-of-the-art MS methods on the GoPro dataset in PSNR with the smallest number of parameters.

Author

Park, Dongwon; Kang, Dong Un; Kim, Jisoo; Chun, Se Young

Published

2019

Language

EN