P
Patrick Lucey
Researcher at Disney Research
Publications - 126
Citations - 7846
Patrick Lucey is an academic researcher from Disney Research. The author has contributed to research in topics: Facial recognition system & Audio-visual speech recognition. The author has an hindex of 31, co-authored 126 publications receiving 6527 citations. Previous affiliations of Patrick Lucey include University of Pittsburgh & Queensland University of Technology.
Papers
More filters
Proceedings ArticleDOI
Using a Free-Parts Representation for Visual Speech Recognition
TL;DR: In this paper, free-parts based representations are used for speaker-independent visual speech recognition, where the position/structure of patches within the mouth image can be relaxed so they can "freely" move to varying extents, hence reducing the influence of the front-end effect.
LipreadingUsing ProfileVersusFrontal Views
TL;DR: This paper investigates extracting visual speech information from the speaker's profile view, and constitutes the first real attempt toattack this problem of automatic speech recognition robustness.
Improved speech reading through a free-parts representation.
Simon Lucey,Patrick Lucey +1 more
TL;DR: This approach additionally equires a modification to traditional techniques employed fo r the estimation of hidden Markov Models (HMMs), whose resultant models the authors currently refer to as free-parts HMMs (FP-HMMs) will be presented on the CUAVE audiovisual speech database.
Patent
System and method for generating trackable video frames from broadcast video
TL;DR: In this article, a system and method of generating trackable frames from a broadcast video feed is presented, where the computing system partitions each frame of the set of video frames into a plurality of clusters and classifies each cluster as trackable or untrackable.
Posted Content
You Cannot Do That Ben Stokes: Dynamically Predicting Shot Type in Cricket Using a Personalized Deep Neural Network.
TL;DR: In this paper, the authors presented a personalized deep neural network approach which can predict the probabilities of where a specific batsman will hit a specific bowler and bowl type, in a specific game-scenario.