TY - GEN
T1 - Latent-Based Adversarial Neural Networks for Facial Affect Estimations
AU - Aspandi, Decky
AU - Mallol-Ragolta, Adria
AU - Schuller, Bjorn
AU - Binefa, Xavier
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/11
Y1 - 2020/11
N2 - There is a growing interest in affective computing research nowadays given its crucial role in bridging humans with computers. This progress has recently been accelerated due to the emergence of bigger dataset. One recent advance in this field is the use of adversarial learning to improve model learning through augmented samples. However, the use of latent features, which is feasible through adversarial learning, is not largely explored, yet. This technique may also improve the performance of affective models, as analogously demonstrated in related fields, such as computer vision. To expand this analysis, in this work, we explore the use of latent features through our proposed adversarial-based networks for valence and arousal recognition in the wild. Specifically, our models operate by aggregating several modalities to our discriminator, which is further conditioned to the extracted latent features by the generator. Our experiments on the recently released SEWA dataset suggest the progressive improvements of our results. Finally, we show our competitive results on the Affective Behavior Analysis in-the-Wild (ABAW) challenge dataset.
AB - There is a growing interest in affective computing research nowadays given its crucial role in bridging humans with computers. This progress has recently been accelerated due to the emergence of bigger dataset. One recent advance in this field is the use of adversarial learning to improve model learning through augmented samples. However, the use of latent features, which is feasible through adversarial learning, is not largely explored, yet. This technique may also improve the performance of affective models, as analogously demonstrated in related fields, such as computer vision. To expand this analysis, in this work, we explore the use of latent features through our proposed adversarial-based networks for valence and arousal recognition in the wild. Specifically, our models operate by aggregating several modalities to our discriminator, which is further conditioned to the extracted latent features by the generator. Our experiments on the recently released SEWA dataset suggest the progressive improvements of our results. Finally, we show our competitive results on the Affective Behavior Analysis in-the-Wild (ABAW) challenge dataset.
KW - Adversarial Networks
KW - Affective Computing
KW - Latent Representation
UR - http://www.scopus.com/inward/record.url?scp=85101461458&partnerID=8YFLogxK
U2 - 10.1109/FG47880.2020.00053
DO - 10.1109/FG47880.2020.00053
M3 - Conference contribution
AN - SCOPUS:85101461458
T3 - Proceedings - 2020 15th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2020
SP - 606
EP - 610
BT - Proceedings - 2020 15th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2020
A2 - Struc, Vitomir
A2 - Gomez-Fernandez, Francisco
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 15th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2020
Y2 - 16 November 2020 through 20 November 2020
ER -