Dr. Muhammad Zeshan Afzal

Prof. Dr. Didier Stricker

Dr. Alain Pagani

Keonna Cunningham

Dr. Oliver Wasenmüller

Dr. Gabriele Bleser

Dr. Bruno Mirbach

Dr. Jason Raphael Rambach

Dr. Bertram Taetz

Mhd Rashed Al Koutayni

Jilliam Maria Diaz Barros

Iuliia Brishtel

Mahdi Chamseddine

Steve Dias da Cruz

Ahmet Firintepe

Sophie Folawiyo

David Michael Fürst

Kamalveerkaur Garewal

Christiano Couto Gava

Leif Eric Goebel

Tewodros Amberbir Habtegebrial

Jigyasa Singh Katrolia

Andreas Kölsch

Onorina Kovalenko

Muhammad Jameel Nawaz Malik

Nareg Minaskan Karabid

Mohammad Minouei

Manthan Pancholi

Engr. Kumail Raza

Dr. Nadia Robertini

María Alejandra Sánchez Marín

Dr. Kripasindhu Sarkar

Alexander Schäfer

Pascal Schneider

Lukas Stefan Staecker

Dr. Vladislav Golyanik

Dr. Aditya Tewari

André Luiz Brandão

Illumination Normalization by Partially Impossible Encoder-Decoder Cost Function

Illumination Normalization by Partially Impossible Encoder-Decoder Cost Function
Steve Dias Da Cruz, Bertram Taetz, Thomas Stifter, Didier Stricker
Proceedings of IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE Winter Conference on Applications of Computer Vision (WACV-2021) January 5-9 Online IEEE 2021 .

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: https://sviro.kl.dfki.de/

Back to the overview