ECCV 2020 - presentation - Fast Video Object Segmentation using the Global Context Module

ECCV 2020

ECCV 2020 - presentation - Fast Video Object Segmentation using the Global Context Module

Aug 19, 2020
|
87 views
|
Details
Presentation on Fast Video Object Segmentation using the Global Context Module, an ECCV 2020 work by Tencent and The University of Hong Kong. Paper link: https://arxiv.org/abs/2001.11243 TL;DR: We proposed a novel memory module for neural networks that effectively and efficiently propagates information through an arbitrarily long video, the global context (GC) module, and achieved both state-of-the-art accuracy and a real-time processing speed on semi-supervised video object segmentation (VOS) for the first time.

Comments
loading...