%0 Conference Proceedings %T Incorporating depth information into few-shot semantic segmentation %+ Equipe VIBOT - VIsion pour la roBOTique [ImViA EA7535 - ERL CNRS 6000] (VIBOT) %+ Informatique, BioInformatique, Systèmes Complexes (IBISC) %+ Joint MSc in VIsion and RoBOTics [VIBOT] (Master VIBOT) %A Zhang, Yifei %A Sidibé, Désiré %A Morel, Olivier %A Meriaudeau, Fabrice %< avec comité de lecture %B 25th International Conference on Pattern Recognition (ICPR 2020) %C Milan, Italy %3 Proc. of the 25th International Conference on Pattern Recognition (ICPR 2020) %P 3582--3588 %8 2021-01-10 %D 2021 %Z Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] %Z Computer Science [cs]/Artificial Intelligence [cs.AI] %Z Computer Science [cs]/Signal and Image ProcessingConference papers %X Few-shot segmentation presents a significant challengefor semantic scene understanding under limited supervision.Namely, this task targets at generalizing the segmentationability of the model to new categories given a few samples.In order to obtain complete scene information, we extend theRGB-centric methods to take advantage of complementary depthinformation. In this paper, we propose a two-stream deep neuralnetwork based on metric learning. Our method, known as RDNet,learns class-specific prototype representations within RGB anddepth embedding spaces, respectively. The learned prototypesprovide effective semantic guidance on the corresponding RGBand depth query image, leading to more accurate performance.Moreover, we build a novel outdoor scene dataset, known asCityscapes-3i, using labeled RGB images and depth imagesfrom the Cityscapes dataset. We also perform ablation studiesto explore the effective use of depth information in few-shotsegmentation tasks. Experiments on Cityscapes-3i show that ourmethod achieves excellent results with visual and complementarygeometric cues from only a few labeled examples. %G English %2 https://univ-evry.hal.science/hal-02887063/document %2 https://univ-evry.hal.science/hal-02887063/file/ICPR_2020_YZ_DS_OM_FM.pdf %L hal-02887063 %U https://univ-evry.hal.science/hal-02887063 %~ UNIV-BOURGOGNE %~ CNRS %~ UNIV-EVRY %~ IBISC %~ UNIV-PARIS-SACLAY %~ UNIV-EVRY-SACLAY %~ IMVIA %~ VIBOT %~ IBISC-SIAM %~ TEST-HALCNRS %~ UNIVERSITE-PARIS-SACLAY %~ GS-ENGINEERING %~ GS-COMPUTER-SCIENCE %~ GS-LIFE-SCIENCES-HEALTH