Instance Semantic Segmentation via Scale-Aware Patch Fusion Network

Autor:	Jingling Zhang, Meijie Wang, Jinfu Yang, Mingai Li
Rok vydání:	2017
Předmět:	Fusion business.industry Computer science ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Segmentation Pattern recognition Artificial intelligence business Scale (map) Generator (mathematics)
Zdroj:	Communications in Computer and Information Science ISBN: 9789811073014 CCCV (2)
DOI:	10.1007/978-981-10-7302-1_43
Popis:	Instance semantic segmentation has already been a promising direction, but many leading approaches are lack of detailed structural information and unable to segment small size objects. In this paper, we present a novel segmentation framework, called Scale-aware Patch Fusion Network (SPF). Our unified end-to-end trainable network consists of three components, namely, multi-scale patch generator, semantic segmentation network and patch fusion algorithm. This patch-based method aggregates information from different scales of patches via fusing local segmentation prediction results. The proposed approach is thus more effective and simple. Experiments on VOC 2012 segmentation val, VOC 2012 SDS val, MS COCO datasets validate the effectiveness of our approach.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::684a183e58c256d52f10e59240253219 https://doi.org/10.1007/978-981-10-7302-1_43 Zobrazit plný text záznamu