Published on Sat Dec 10 2016

Generalized Deep Image to Image Regression

Venkataraman Santhanam, Vlad I. Morariu, Larry S. Davis

Recursively Branched Deconvolutional Network (RBDN) develops a cheap multi-context image representation very early on. RBDN is fully convolutional and can handle variable sized images duringference.

0
0
0
Abstract

We present a Deep Convolutional Neural Network architecture which serves as a generic image-to-image regressor that can be trained end-to-end without any further machinery. Our proposed architecture: the Recursively Branched Deconvolutional Network (RBDN) develops a cheap multi-context image representation very early on using an efficient recursive branching scheme with extensive parameter sharing and learnable upsampling. This multi-context representation is subjected to a highly non-linear locality preserving transformation by the remainder of our network comprising of a series of convolutions/deconvolutions without any spatial downsampling. The RBDN architecture is fully convolutional and can handle variable sized images during inference. We provide qualitative/quantitative results on diverse tasks: relighting, denoising and colorization and show that our proposed RBDN architecture obtains comparable results to the state-of-the-art on each of these tasks when used off-the-shelf without any post processing or task-specific architectural modifications.

Sat Sep 02 2017
Machine Learning
Fast Image Processing with Fully-Convolutional Networks
The approach uses a fully-convolutional network that is trained on input-output pairs that demonstrate the operator's action. After training, the original operator need not be run at all. Experiments demonstrate that the presented approach is significantly more accurate than prior approximation schemes.
0
0
0
Fri Nov 17 2017
Machine Learning
xUnit: Learning a Spatial Activation Function for Efficient Image Restoration
In recent years, deep neural networks (DNNs) have achieved unprecedented performance in many low-level vision tasks. To make DNNs implementable on platforms with limited resources, it is necessary to weaken the tradeoff between performance and efficiency. In this paper, we propose a new
0
0
0
Tue Nov 27 2018
Computer Vision
Dense xUnit Networks
Deep net architectures have constantly evolved over the past few years, leading to significant advancements in a wide array of computer vision tasks. Many applications also require a low computational load and limited memory footprint. To date, efficiency has typically been achieved either by architectural choices at the macro level or modifications at the individual layers.
0
0
0
Fri Apr 07 2017
Computer Vision
Convolutional Neural Pyramid for Image Processing
We propose a principled convolutional neural pyramid (CNP) framework for general low-level vision and image processing tasks. Extra benefit includes adaptive network depth and progressive upsampling for quasi-realtime testing on VGA-size input.
0
0
0
Tue Oct 02 2018
Machine Learning
Deep Decoder: Concise Image Representations from Untrained Non-convolutional Networks
Deep neural networks have become highly effective tools for compressing images. This success can be attributed in part to their ability to represent and generate natural images well. In this paper, we propose an untrained simple image model, called the deep decoder, which is a deep neural network.
0
0
0
Thu Feb 18 2021
Computer Vision
DSRN: an Efficient Deep Network for Image Relighting
Deep image relighting allows automatic photo enhancement by illumination-specific retouching. Custom and natural lighting conditions can be emulated in images of the scene. Extraordinary capabilities of the deep learning framework can be utilized for such purpose.
0
0
0