期刊
PATTERN RECOGNITION
卷 116, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.patcog.2021.107953
关键词
VD-ZSAR; Ventral & Dorsal Stream Theory; Nonlinear similarity metric learning; mechanism
资金
- Shenzhen Science and Technology Foundation [JCYJ20170816093943197]
- Science and Technology Program of Guangzhou, China [202002030263]
- Guangdong Basic and Applied Basic Research Foundation [2020A1515110997]
The proposed VD-ZSAR method extracts nonredundant visual features based on the Ventral & Dorsal Stream Theory, and learns the correlation between actions through a visual-semantic joint embedding space. Experimental results demonstrate the favorable performance of the approach across multiple datasets.
A B S T R A C T Most Zero-Shot Action Recognition (ZSAR) methods establish visual-semantic joint embedding space, which is based on commonly used visual features and semantic embeddings, to learn the correlation between actions. Nevertheless, extracting visual features without structural guidance would lead to sparse video features, which reflect the correlation of actions, fall into oblivion. Based on the Ventral & Dorsal Stream Theory (VD), we propose a VD-ZSAR method to extract irredundant visual feature, which can relieve relation ambiguity caused by redundant visual feature. And a visual-semantic joint embedding space is learned by combining nonredundant visual space with semantic space. Specifically, visual space is constructed by the motion cues perceived by Dorsal Stream, and the object cues perceived by Ventral Stream. Semantic space is constructed by sentence-to-vector generator. The visual-semantic joint embedding space is built by a nonlinear similarity metric learning mechanism, which can better implicitly reflect the correlation between actions. Extensive experiments on the Olympic, HDMB51 and UCF101 datasets validate the favorable performance of our proposed approach. (c) 2021 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据