DooDLeNet: Double DeepLab Enhanced Feature Fusion for Thermal-Color Semantic Segmentation

Oriel Frigo, Lucien Martin-Gaffe, Catherine Wacongne; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2022, pp. 3021-3029

Abstract

In this paper we present a new approach for feature fusion between RGB and LWIR Thermal images for the task of semantic segmentation for driving perception. We propose the DooDLeNet, a double DeepLab architecture with specialized encoder-decoders for thermal and color modalities and a shared decoder for final segmentation. We combine two strategies for feature fusion: confidence weighting and correlation weighting. We report state-of-the-art mean IoU results on MF dataset.

Related Material

[pdf]
[bibtex]
@InProceedings{Frigo_2022_CVPR, author = {Frigo, Oriel and Martin-Gaffe, Lucien and Wacongne, Catherine}, title = {DooDLeNet: Double DeepLab Enhanced Feature Fusion for Thermal-Color Semantic Segmentation}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2022}, pages = {3021-3029} }