Learning-based Image Enhancement for Visual Odometry in Challenging HDR Environments

TL;DR

This paper introduces a learning-based image enhancement method using deep neural networks to improve visual odometry performance in challenging HDR environments, emphasizing real-time applicability and sequence consistency.

Contribution

It proposes a novel deep learning framework with both LSTM and CNN architectures to enhance image sequences for robust VO under difficult lighting conditions.

Findings

Enhanced VO accuracy in HDR environments.

LSTM-based sequences show improved temporal consistency.

Reduced-size CNN enables real-time processing.

Abstract

One of the main open challenges in visual odometry (VO) is the robustness to difficult illumination conditions or high dynamic range (HDR) environments. The main difficulties in these situations come from both the limitations of the sensors and the inability to perform a successful tracking of interest points because of the bold assumptions in VO, such as brightness constancy. We address this problem from a deep learning perspective, for which we first fine-tune a Deep Neural Network (DNN) with the purpose of obtaining enhanced representations of the sequences for VO. Then, we demonstrate how the insertion of Long Short Term Memory (LSTM) allows us to obtain temporally consistent sequences, as the estimation depends on previous states. However, the use of very deep networks does not allow the insertion into a real-time VO framework; therefore, we also propose a Convolutional Neural Network (CNN) of reduced size capable of performing faster. Finally, we validate the enhanced representations by evaluating the sequences produced by the two architectures in several state-of-art VO algorithms, such as ORB-SLAM and DSO.