Summary of Sun Team’s Contribution to Abaw 2024 Competition: Audio-visual Valence-arousal Estimation and Expression Recognition, by Denis Dresvyanskiy et al.
SUN Team’s Contribution to ABAW 2024 Competition: Audio-visual Valence-Arousal Estimation and Expression Recognition
by Denis Dresvyanskiy, Maxim Markitantov, Jiawei Yu, Peitong Li, Heysem Kaya, Alexey Karpov
First submitted to arxiv on: 19 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A deep learning approach is developed for recognizing emotions in videos and audio recordings, focusing on real-world scenarios rather than controlled laboratory settings. The method combines Convolutional Neural Networks (CNNs) for visual features and Public Dimensional Emotion Model (PDEM) for audio cues. Multiple temporal modeling strategies are evaluated to determine the most effective way to combine these modalities. The approach is tested on the AffWild2 dataset, which is used in the Affective Behavior Analysis in-the-Wild 2024 (ABAW’24) challenge. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way has been found to recognize emotions in videos and audio recordings that works well in real-life situations. This method uses special kinds of artificial intelligence called Convolutional Neural Networks (CNNs) for the visual part and another type called Public Dimensional Emotion Model (PDEM) for the audio part. The researchers tested different ways of combining these parts to see which one worked best. They used a dataset called AffWild2, which is important for understanding how emotions are expressed in real-life situations. |
Keywords
* Artificial intelligence * Deep learning
