Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226156701> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4226156701 abstract "Automatically understanding and recognising human affective states using images and computer vision can improve human-computer and human-robot interaction. However, privacy has become an issue of great concern, as the identities of people used to train affective models can be exposed in the process. For instance, malicious individuals could exploit images from users and assume their identities. In addition, affect recognition using images can lead to discriminatory and algorithmic bias, as certain information such as race, gender, and age could be assumed based on facial features. Possible solutions to protect the privacy of users and avoid misuse of their identities are to: (1) extract anonymised facial features, namely action units (AU) from a database of images, discard the images and use AUs for processing and training, and (2) federated learning (FL) i.e. process raw images in users' local machines (local processing) and send the locally trained models to the main processing machine for aggregation (central processing). In this paper, we propose a two-level deep learning architecture for affect recognition that uses AUs in level 1 and FL in level 2 to protect users' identities. The architecture consists of recurrent neural networks to capture the temporal relationships amongst the features and predict valence and arousal affective states. In our experiments, we evaluate the performance of our privacy-preserving architecture using different variations of recurrent neural networks on RECOLA, a comprehensive multimodal affective database. Our results show state-of-the-art performance of $0.426$ for valence and $0.401$ for arousal using the Concordance Correlation Coefficient evaluation metric, demonstrating the feasibility of developing models for affect recognition that are both accurate and ensure privacy." @default.
- W4226156701 created "2022-05-05" @default.
- W4226156701 creator A5018748194 @default.
- W4226156701 creator A5024981509 @default.
- W4226156701 creator A5039053107 @default.
- W4226156701 creator A5051446557 @default.
- W4226156701 creator A5069160503 @default.
- W4226156701 date "2021-11-14" @default.
- W4226156701 modified "2023-09-28" @default.
- W4226156701 title "Towards Privacy-Preserving Affect Recognition: A Two-Level Deep Learning Architecture" @default.
- W4226156701 hasPublicationYear "2021" @default.
- W4226156701 type Work @default.
- W4226156701 citedByCount "0" @default.
- W4226156701 crossrefType "posted-content" @default.
- W4226156701 hasAuthorship W4226156701A5018748194 @default.
- W4226156701 hasAuthorship W4226156701A5024981509 @default.
- W4226156701 hasAuthorship W4226156701A5039053107 @default.
- W4226156701 hasAuthorship W4226156701A5051446557 @default.
- W4226156701 hasAuthorship W4226156701A5069160503 @default.
- W4226156701 hasBestOaLocation W42261567011 @default.
- W4226156701 hasConcept C107457646 @default.
- W4226156701 hasConcept C119857082 @default.
- W4226156701 hasConcept C121332964 @default.
- W4226156701 hasConcept C123657996 @default.
- W4226156701 hasConcept C142362112 @default.
- W4226156701 hasConcept C153349607 @default.
- W4226156701 hasConcept C154945302 @default.
- W4226156701 hasConcept C15744967 @default.
- W4226156701 hasConcept C165696696 @default.
- W4226156701 hasConcept C168900304 @default.
- W4226156701 hasConcept C195704467 @default.
- W4226156701 hasConcept C2776035688 @default.
- W4226156701 hasConcept C38652104 @default.
- W4226156701 hasConcept C41008148 @default.
- W4226156701 hasConcept C46312422 @default.
- W4226156701 hasConcept C50644808 @default.
- W4226156701 hasConcept C62520636 @default.
- W4226156701 hasConceptScore W4226156701C107457646 @default.
- W4226156701 hasConceptScore W4226156701C119857082 @default.
- W4226156701 hasConceptScore W4226156701C121332964 @default.
- W4226156701 hasConceptScore W4226156701C123657996 @default.
- W4226156701 hasConceptScore W4226156701C142362112 @default.
- W4226156701 hasConceptScore W4226156701C153349607 @default.
- W4226156701 hasConceptScore W4226156701C154945302 @default.
- W4226156701 hasConceptScore W4226156701C15744967 @default.
- W4226156701 hasConceptScore W4226156701C165696696 @default.
- W4226156701 hasConceptScore W4226156701C168900304 @default.
- W4226156701 hasConceptScore W4226156701C195704467 @default.
- W4226156701 hasConceptScore W4226156701C2776035688 @default.
- W4226156701 hasConceptScore W4226156701C38652104 @default.
- W4226156701 hasConceptScore W4226156701C41008148 @default.
- W4226156701 hasConceptScore W4226156701C46312422 @default.
- W4226156701 hasConceptScore W4226156701C50644808 @default.
- W4226156701 hasConceptScore W4226156701C62520636 @default.
- W4226156701 hasLocation W42261567011 @default.
- W4226156701 hasOpenAccess W4226156701 @default.
- W4226156701 hasPrimaryLocation W42261567011 @default.
- W4226156701 hasRelatedWork W10279623 @default.
- W4226156701 hasRelatedWork W10372156 @default.
- W4226156701 hasRelatedWork W1319706 @default.
- W4226156701 hasRelatedWork W1383942 @default.
- W4226156701 hasRelatedWork W15108508 @default.
- W4226156701 hasRelatedWork W15696342 @default.
- W4226156701 hasRelatedWork W17729602 @default.
- W4226156701 hasRelatedWork W2904021 @default.
- W4226156701 hasRelatedWork W349990 @default.
- W4226156701 hasRelatedWork W8694306 @default.
- W4226156701 isParatext "false" @default.
- W4226156701 isRetracted "false" @default.
- W4226156701 workType "article" @default.