AffWild Net and Aff-Wild Database
Alvertos Benroumpi, Dimitrios Kollias

TL;DR
This paper introduces AffWild Net and the Aff-Wild database for 'in the wild' emotion recognition, utilizing deep learning models like CNNs and a novel Regression GAN to analyze real-world images for valence and arousal.
Contribution
It presents a new large-scale dataset and a novel Regression GAN model capable of unsupervised and supervised learning for emotion recognition in real-world images.
Findings
Created a dataset with over 507,000 frames from online videos.
Implemented a CNN-M based model for data usability testing.
Developed a Regression GAN that predicts valence and arousal for real and fake images.
Abstract
Emotions recognition is the task of recognizing people's emotions. Usually it is achieved by analyzing expression of peoples faces. There are two ways for representing emotions: The categorical approach and the dimensional approach by using valence and arousal values. Valence shows how negative or positive an emotion is and arousal shows how much it is activated. Recent deep learning models, that have to do with emotions recognition, are using the second approach, valence and arousal. Moreover, a more interesting concept, which is useful in real life is the "in the wild" emotions recognition. "In the wild" means that the images analyzed for the recognition task, come from from real life sources(online videos, online photos, etc.) and not from staged experiments. So, they introduce unpredictable situations in the images, that have to be modeled. The purpose of this project is to study…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsEmotion and Mood Recognition · Face recognition and analysis · Face and Expression Recognition
MethodsConvolution · Dogecoin Customer Service Number +1-833-534-1729
