-
Deep Networks for Image Super-Resolution with Sparse Prior
Deep learning techniques have been successfully applied in many areas of...
07/31/2015 ∙ by Zhaowen Wang, et al. ∙
0
∙
share
read it
-
Hierarchical Back Projection Network for Image Super-Resolution
Deep learning based single image super-resolution methods use a large nu...
06/17/2019 ∙ by Zhi-Song Liu, et al. ∙
2
∙
share
read it
-
Fast and Accurate Image Super Resolution by Deep CNN with Skip Connection and Network in Network
We propose a highly efficient and faster Single Image Super-Resolution (...
07/18/2017 ∙ by Jin Yamanaka, et al. ∙
0
∙
share
read it
-
CT-image Super Resolution Using 3D Convolutional Neural Network
Computed Tomography (CT) imaging technique is widely used in geological ...
06/24/2018 ∙ by Yukai Wang, et al. ∙
4
∙
share
read it
-
Deep learning-based super-resolution in coherent imaging systems
We present a deep learning framework based on a generative adversarial n...
10/15/2018 ∙ by Tairan Liu, et al. ∙
0
∙
share
read it
-
Geometry-Aware Neighborhood Search for Learning Local Models for Image Reconstruction
Local learning of sparse image models has proven to be very effective to...
05/06/2015 ∙ by Julio Cesar Ferreira, et al. ∙
0
∙
share
read it
-
Image Reconstruction with Predictive Filter Flow
We propose a simple, interpretable framework for solving a wide range of...
11/28/2018 ∙ by Shu Kong, et al. ∙
12
∙
share
read it
Controlling Neural Networks via Energy Dissipation
04/05/2019 ∙ by Michael Moeller, et al. ∙
0
∙
share
The last decade has shown a tremendous success in solving various computer vision problems with the help of deep learning techniques. Lately, many works have demonstrated that learning-based approaches with suitable network architectures even exhibit superior performance for the solution of (ill-posed) image reconstruction problems such as deblurring, super-resolution, or medical image reconstruction. The drawback of purely learning-based methods, however, is that they cannot provide provable guarantees for the trained network to follow a given data formation process during inference. In this work we propose energy dissipating networks that iteratively compute a descent direction with respect to a given cost function or energy at the currently estimated reconstruction. Therefore, an adaptive step size rule such as a line-search, along with a suitable number of iterations can guarantee the reconstruction to follow a given data formation model encoded in the energy to arbitrary precision, and hence control the model's behavior even during test time. We prove that under standard assumptions, descent using the direction predicted by the network converges (linearly) to the global minimum of the energy. We illustrate the effectiveness of the proposed approach in experiments on single image super resolution and computed tomography (CT) reconstruction, and further illustrate extensions to convex feasibility problems.
READ FULL TEXT