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

Oriel Frigo, Lucien Martin-Gaffe, Catherine Wacongne

20222022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)28 citationsDOI

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 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 the MF dataset [1].

Topics & Concepts

WeightingComputer scienceFeature (linguistics)Artificial intelligenceSegmentationPattern recognition (psychology)EncoderPairwise comparisonFusionRGB color modelTask (project management)Operating systemLinguisticsPhilosophyEconomicsMedicineRadiologyManagementAdvanced Neural Network ApplicationsVideo Surveillance and Tracking MethodsImage Enhancement Techniques