Video Object Segmentation Using Kernelized Memory Network With Multiple Kernels

Autor:	Hongje Seong, Junhyuk Hyun, Euntai Kim
Rok vydání:	2023
Předmět:	Computational Theory and Mathematics Artificial Intelligence Applied Mathematics Computer Vision and Pattern Recognition Software
Zdroj:	IEEE Transactions on Pattern Analysis and Machine Intelligence. 45:2595-2612
ISSN:	1939-3539 0162-8828
DOI:	10.1109/tpami.2022.3163375
Popis:	Semi-supervised video object segmentation (VOS) is to predict the segment of a target object in a video when a ground truth segmentation mask for the target is given in the first frame. Recently, space-time memory networks (STM) have received significant attention as a promising approach for semi-supervised VOS. However, an important point has been overlooked in applying STM to VOS: The solution (=STM) is non-local, but the problem (=VOS) is predominantly local. To solve this mismatch between STM and VOS, we propose new VOS networks called kernelized memory network (KMN) and KMN with multiple kernels (KMN
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8f607dcf78d834be0bdd8c3545364415 https://doi.org/10.1109/tpami.2022.3163375 Zobrazit plný text záznamu