A
Andrea Cavallaro
Researcher at Queen Mary University of London
Publications - 366
Citations - 10738
Andrea Cavallaro is an academic researcher from Queen Mary University of London. The author has contributed to research in topics: Video tracking & Object detection. The author has an hindex of 46, co-authored 345 publications receiving 8945 citations. Previous affiliations of Andrea Cavallaro include Tel Aviv University & Dalhousie University.
Papers
More filters
Proceedings ArticleDOI
Macro-cuboïd based probabilistic matching for lip-reading digits
TL;DR: The evaluation shows that the proposed approach outperforms the others in recognition accuracy and is robust in coping with variations in probe sequences.
Proceedings ArticleDOI
Improving Filling Level Classification with Adversarial Training
TL;DR: It is shown that transfer learning with adversarial training in the source domain consistently improves the classification accuracy on the test set and limits the overfitting of the classifier to specific features of the training data.
Journal ArticleDOI
Audio-visual tracking of concurrent speakers
TL;DR: A tracker that builds on generative and discriminative audio-visual likelihood models formulated in a particle filtering framework that outperforms the uni-modal trackers and the state-of theart approaches both in 3D and on the image plane.
Proceedings ArticleDOI
Region Segmentation and Feature Point Extraction on 3D Faces using a Point Distribution Model
P. Nair,Andrea Cavallaro +1 more
TL;DR: A novel approach to accurately detect landmarks and segment regions on face meshes without the use of texture, pose or orientation information is presented, based on a 3D point distribution model that is fitted to the region of interest using candidate vertices extracted from low-level feature maps.
Proceedings ArticleDOI
Evaluation of on-line quality estimators for object tracking
TL;DR: A taxonomy is proposed and a comparative evaluation of online quality estimators for video object tracking shows that the Observation Likelihood measure is an appropriate quality measure for overall tracking performance evaluation, while the Template Inverse Matching measure is appropriate to detect the start and the end instants of tracking failures.