Exploiting sparsity and co-occurrence structure for action unit recognition

By Yale Song, Daniel J. McDuff, Deepak Vasisht, Ashish Kapoor. 2015

64 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Experimental and Cognitive Psychology, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Maja Pantic
2. Michel François Valstar
3. Dacheng Tao
4. Luc Van Gool
5. Qiang Ji
6. Yale Song
7. Brais Martínez
8. Michal Drozdzal

In Experimental and Cognitive Psychology

1. Maja Pantic
2. Michel François Valstar
3. Brais Martínez
4. Bihan Jiang
5. Yale Song
6. Qiang Ji
7. Tengfei Song
8. Yuru Wang

In Signal Processing

1. Qiang Ji
2. Michel François Valstar
3. Maja Pantic
4. Brais Martínez
5. Tengfei Song
6. Adriana Romero
7. Mingli Song
8. Yuncheng Li

Citations in the corpus

• Elucidating image-to-set prediction: An analysis of models, losses and datasets (2019, background): “Traditionally, the problem has been tackled from many different perspectives, from decomposing the problem into independent binary predictions [33, 58] or modeling label correlations [1, 43, 28], to exploiting priors such as label noise and sparsity [16, 19, 45, 54, 59, 2].”
• Knowledge Augmented Deep Neural Networks for Joint Facial Expression and Action Unit Recognition (2020, background): “Moreover, graphical models attract much attention due to their power of modeling probabilistic dependencies([47, 43, 5]).”
• Automatic Analysis of Facial Actions: A Survey (2019, methodology): “Finally, [166] employed a graphical model, a variant of the Bayesian compressed sensing framework, capable of grouping AU (where an AU can be on more than one group), and imposing sparsity so few AU can be active at a time.”
• CU-Net: Component Unmixing Network for Textile Fiber Identification (2019, background): “These approaches usually exploit distinctive properties of multilabel classification, such as label sparsity (Song et al. 2015; Hsu et al. 2009; Kapoor et al. 2012; Yang et al. 2014, 2018; Zhou and Tao 2012; Zhang et al. 2015), label dependency (Shu et al. 2015; Alessandro et al. 2013; Liu and…”
• Improving Pairwise Ranking for Multi-label Image Classification (2017, methodology): “We propose a new pairwise ranking loss function and a per-class threshold estimation method in a unified framework, improving existing ranking-based approaches in a principled manner. label dependency [1, 25], label sparsity [10, 12, 27], and label noise [33, 39].”