Submitted by Marin Orsic.

Submission data

Full nameSwiftNet pyr RN152 RVC
DescriptionSwiftNet pyr RN152 submission to 2020 RVC
Publication titleIn Defense of Pre-trained ImageNet Architectures for Real-time Semantic Segmentation of Road-driving Images
Publication authorsMarin Oršić, Ivan Krešo, Petra Bevandić, Siniša Šegvić
Publication venueCVPR 2019
Publication URLhttps://openaccess.thecvf.com/content_CVPR_2019/papers/Orsic_In_Defense_of_Pre-Trained_ImageNet_Architectures_for_Real-Time_Semantic_Segmentation_CVPR_2019_paper.pdf
Input Data TypesUses Color        Uses 2D
Programming language(s)Python/PyTorch/CUDA
HardwareTesla V100
Websitehttps://github.com/orsic/swiftnet
Source code or download URLhttps://github.com/orsic/swiftnet
Submission creation date14 Aug, 2020
Last edited14 Aug, 2020

2D semantic label results

Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
copyleft0.5460.5720.6630.6380.5180.2980.3660.6330.5100.4460.8640.2960.2670.5420.3460.7040.5750.4310.8530.7660.630