Loss Max-Pooling for Semantic Image Segmentation
Authors:
- Samuel Rota BuloSBSamuel Rota Bulo
- Gerhard NeuholdGNGerhard Neuhold
- Peter KontschiederPKPeter Kontschieder
- 1 July 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636919,p. 7082-7091
Abstract
We introduce a novel loss max-pooling concept for handling imbalanced training data distributions, applicable as alternative loss layer in the context of deep neural networks for semantic image segmentation. Most real-world semantic segmentation datasets exhibit long tail distributions with few object categories comprising the majority of data and consequently biasing the classifiers towards them. Our method adaptively re-weights the contributions of each pixel based on their observed losses, targeting under-performing classification results as often encountered for under-represented object classes. Our approach goes beyond conventional cost-sensitive learning attempts through adaptive considerations that allow us to indirectly address both, inter- and intra-class imbalances. We provide a theoretical justification of our approach, complementary to experimental analyses on benchmark datasets. In our experiments on the Cityscapes and Pascal VOC 2012 segmentation datasets we find consistently improved results, demonstrating the efficacy of our approach.Keywords
All Related Versions
This publication has 26 references indexed in Scilit:
- Learning to segment under various forms of weak supervision2015
- The Pascal Visual Object Classes Challenge: A RetrospectiveInternational Journal of Computer Vision, 2014
- Optimal Decisions from Probabilistic Models: The Intersection-over-Union Case2014
- Tell Me What You See and I Will Show You Where It Is2014
- Optimizing Complex Loss Functions in Structured Prediction2010
- Classification of Imbalanced Data by Combining the Complementary Neural Network and SMOTE Algorithm2010
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009
- Learning to Localize Objects with Structured Output Regression2008
- Extreme re-balancing for SVMsACM SIGKDD Explorations Newsletter, 2004
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998