Semantic Segmentation via Multi-task, Multi-domain Learning

Abstract : We present an approach that leverages multiple datasets possibly annotated using different classes to improve the semantic segmentation accuracy on each individual dataset. We propose a new selective loss function that can be integrated into deep networks to exploit training data coming from multiple datasets with possibly different tasks (e.g., different label-sets). We show how the gradient-reversal approach for domain adaptation can be used in this setup. Thorought experiments on semantic segmentation applications show the relevance of our approach.
Complete list of metadatas

Cited literature [26 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01376998
Contributor : Damien Fourure <>
Submitted on : Thursday, October 6, 2016 - 10:18:33 AM
Last modification on : Tuesday, February 26, 2019 - 4:35:37 PM
Long-term archiving on : Saturday, January 7, 2017 - 1:01:06 PM

File

Semantic_Segmentation_via_Mult...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01376998, version 1

Citation

Damien Fourure, Rémi Emonet, Elisa Fromont, Damien Muselet, Alain Trémeau, et al.. Semantic Segmentation via Multi-task, Multi-domain Learning. S+SSPR 2016 The joint IAPR International Workshops on Structural and Syntactic Pattern Recognition (SSPR 2016) and Statistical Techniques in Pattern Recognition (SPR 2016) , Nov 2016, Merida, Mexico. ⟨hal-01376998⟩

Share

Metrics

Record views

429

Files downloads

546