Continual Online Learning For Unconstrained Facial Landmark Detection And Tracking
Lead Research Organisation:
University of Nottingham
Department Name: School of Computer Science
Abstract
Facial landmark localization is of profound interest in computer vision since it is critical in solving many key affective computing problems such as face recognition, expression recognition, emotion detection etc. Face landmark alignment, which is all about localization of key facial structures such as eyes, nose, mouth etc., is quite challenging in nature owing to wide range of face appearance variations. This is mainly due to various head poses, environmental lighting conditions, external occlusions, and imaging sensor noise patterns. Most important challenge in building the landmark detection and tracking solution stems from the fact that it is difficult to train a machine learning model with training data that fully spans all the aforementioned attribute space. Hence by enabling the model to learn incrementally from test time environment, this challenge can be addressed to a significant extent. However, incremental online learning often results in a phenomenon known as catastrophic interference i.e. when the model learns new information it starts forgetting the previously learned information. Continual learning approaches are demonstrated to be capable of handling this interference effectively in case of connectionist learning procedures. In this proposal the possibility of employing the continual learning methods to address the catastrophic forgetting phenomenon during incremental training of landmark detection and tracking models is discussed.
Publications
Tellamekala M
(2022)
Modelling Stochastic Context of Audio-Visual Expressive Behaviour with Affective Processes
in IEEE Transactions on Affective Computing
Tellamekala M
(2022)
Dimensional Affect Uncertainty Modelling for Apparent Personality Recognition
in IEEE Transactions on Affective Computing
Tellamekala M
(2019)
Temporally Coherent Visual Representations for Dimensional Affect Recognition
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
EP/R512059/1 | 01/10/2017 | 31/03/2023 | |||
2159382 | Studentship | EP/R512059/1 | 01/10/2018 | 30/09/2022 | Mani Tellamekala |
Description | PhD Tech Talk-Facebook London |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Postgraduate students |
Results and Impact | In this Open House event at Facebook London, I have presented a poster on my research work related to affect recognition. It provided an opportunity to discuss my already published conference paper with Ph.D. students from other universities in the UK. As a result of attending this event, I have gotten some good feedback on my work that helped to shape my later work. |
Year(s) Of Engagement Activity | 2019 |
URL | https://phdopenhouselon19.splashthat.com/ |