Cityscapes-Panoptic-Parts and PASCAL-Panoptic-Parts datasets for Scene Understanding
2020-04-16Code Available1· sign in to hype
Panagiotis Meletis, Xiaoxiao Wen, Chenyang Lu, Daan de Geus, Gijs Dubbelman
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/tue-mps/panoptic_partsOfficialIn papertf★ 105
- github.com/pmeletis/panoptic_partspytorch★ 105
- github.com/chufengt/Visual-Recognition-by-Requestpytorch★ 43
- github.com/chufengt/ViRReqpytorch★ 43
Abstract
In this technical report, we present two novel datasets for image scene understanding. Both datasets have annotations compatible with panoptic segmentation and additionally they have part-level labels for selected semantic classes. This report describes the format of the two datasets, the annotation protocols, the merging strategies, and presents the datasets statistics. The datasets labels together with code for processing and visualization will be published at https://github.com/tue-mps/panoptic_parts.