MMAct: A Large-Scale Dataset for Cross Modal Human Action Understanding

• Published in: Proceedings / IEEE International Conference on Computer Vision pp. 8657 - 8666
• Main Authors: Kong, Quan, Wu, Ziming, Deng, Ziwei, Klinkigt, Martin, Tong, Bin, Murakami, Tomokazu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2019
• Subjects: Adaptation models; Cameras; Gyroscopes; Sensors; Task analysis; Three-dimensional displays; Videos

Unlike vision modalities, body-worn sensors or passive sensing can avoid the failure of action understanding in vision related challenges, e.g. occlusion and appearance variation. However, a standard large-scale dataset does not exist, in which different types of modalities across vision and sensors...