S
Stephen J. Maybank
Researcher at Birkbeck, University of London
Publications - 172
Citations - 17723
Stephen J. Maybank is an academic researcher from Birkbeck, University of London. The author has contributed to research in topics: Video tracking & Motion estimation. The author has an hindex of 50, co-authored 166 publications receiving 15225 citations. Previous affiliations of Stephen J. Maybank include University of Oxford & University of Reading.
Papers
More filters
Journal ArticleDOI
STA-CNN: Convolutional Spatial-Temporal Attention Learning for Action Recognition
TL;DR: A Spatial-Temporal Attentive Convolutional Neural Network (STA-CNN) which selects the discriminative temporal segments and focuses on the informative spatial regions automatically and achieves the state-of-the-art performance on two of the most challenging datasets, UCF-101 and HMDB-51.
Book ChapterDOI
Spatiotemporal attacks for embodied agents
Aishan Liu,Tairan Huang,Xianglong Liu,Yitao Xu,Yuqing Ma,Xinyun Chen,Stephen J. Maybank,Dacheng Tao +7 more
TL;DR: This work generates spatiotemporal perturbations to form 3D adversarial examples, which exploit the interaction history in both the temporal and spatial dimensions and adversarially perturb the physical properties of the contextual objects that appeared in the most important scene views.
Journal ArticleDOI
Occlusion Reasoning for Tracking Multiple People
TL;DR: The problem of tracking and occlusion reasoning for more than two people is formulated mathematically, and a solution is proposed based on particle filtering.
Journal ArticleDOI
Efficient Clustering Aggregation Based on Data Fragments
TL;DR: It is proved that clustering aggregation can be performed directly on data fragments under two widely used goodness measures for clustering aggregations taken from the literature.
Journal ArticleDOI
Image Classification Using Multiscale Information Fusion Based on Saliency Driven Nonlinear Diffusion Filtering
TL;DR: The algorithm emphasizes the foreground features, which are important for image classification, and preserves or even enhances semantically important structures in the foreground, and inhibits and smoothes clutter in the background.