Illumination Normalization by Partially Impossible Encoder-Decoder Cost Function

TL;DR

This paper proposes a novel encoder-decoder cost function for image normalization that emphasizes salient features by ignoring environmental and illumination variations, improving robustness in computer vision tasks.

Contribution

It introduces a partially impossible reconstruction loss combined with triplet loss and nearest neighbor search to enhance generalization across different lighting and environmental conditions.

Findings

Effective normalization of images under varying illumination.

Improved robustness of models to unseen environmental conditions.

Validated on three public datasets and a new synthetic automotive dataset.

Abstract

Images recorded during the lifetime of computer vision based systems undergo a wide range of illumination and environmental conditions affecting the reliability of previously trained machine learning models. Image normalization is hence a valuable preprocessing component to enhance the models' robustness. To this end, we introduce a new strategy for the cost function formulation of encoder-decoder networks to average out all the unimportant information in the input images (e.g. environmental features and illumination changes) to focus on the reconstruction of the salient features (e.g. class instances). Our method exploits the availability of identical sceneries under different illumination and environmental conditions for which we formulate a partially impossible reconstruction target: the input image will not convey enough information to reconstruct the target in its entirety. Its applicability is assessed on three publicly available datasets. We combine the triplet loss as a regularizer in the latent space representation and a nearest neighbour search to improve the generalization to unseen illuminations and class instances. The importance of the aforementioned post-processing is highlighted on an automotive application. To this end, we release a synthetic dataset of sceneries from three different passenger compartments where each scenery is rendered under ten different illumination and environmental conditions: see https://sviro.kl.dfki.de