Understanding the context of complex and cluttered scenes is a challenging problem for semantic segmentation . It is difficult to model the context without prior and additional supervision because the scene’s factors, such as the scale, shape, and appearance of objects, vary considerably in the scene . To solve this, we propose to learn the structures of objects and thehierarchy among objects because context is based on these intrinsic properties . In the experiments, we confirmed that our proposed method achieves state-of-the-art performance in PASCAL Context . In this study, we design novel novel hierarchical, contextual, and multiscalepyramidal representations to capture the properties from an input image. In the experiment, we confirm that our proposal method achieves that our . proposed method achieved .

Author(s) : Hiroaki Aizawa, Yukihiro Domae, Kunihito Kato

Links : PDF - Abstract

Code :
Coursera

Keywords : context - method - objects - hierarchical - segmentation -

Leave a Reply

Your email address will not be published. Required fields are marked *