Facial Expression Intensity Estimation Using Ordinal Information

Zhao, Rui; Gan, Quan; Wang, Shangfei; Ji, Qiang

doi:10.1109/cvpr.2016.377

Cited by 82 publications

(93 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Methods MAE↓ MSE↓ PCC↑ our proposed method 1 0.401 0.742 0.643 our proposed method 2 0.334 0.626 0.804 smooth L1 + L1 center loss [3] 0.456 0.804 0.651 OSVR-L1 [21] 1.025 N/A 0.600 OSVR-L2 [21] 0.810 N/A 0.601 RCR [13] N/A 1.54 0.65 Table 1. Performance of our proposed methods and related works on the UNBC-McMaster Shoulder-Pain dataset for the estimation of pain intensity.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Saliency Supervision: An Intuitive and Effective Approach for Pain Intensity Regression

Zhu

Zhao

2018

Neural Information Processing

View full text Add to dashboard Cite

Getting pain intensity from face images is an important problem in autonomous nursing systems. However, due to the limitation in data sources and the subjectiveness in pain intensity values, it is hard to adopt modern deep neural networks for this problem without domain-specific auxiliary design. Inspired by human vision priori, we propose a novel approach called saliency supervision, where we directly regularize deep networks to focus on facial area that is discriminative for pain regression. Through alternative training between saliency supervision and global loss, our method can learn sparse and robust features, which is proved helpful for pain intensity regression. We verified saliency supervision with face-verification network backbone [1] on the widely-used UNBC-McMaster Shoulder-Pain [2] dataset, and achieved state-of-art performance without bells and whistles. Our saliency supervision is intuitive in spirit, yet effective in performance. We believe such saliency supervision is essential in dealing with ill-posed datasets, and has potential in a wide range of vision tasks.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Methods wMAE↓ wMSE↓ our proposed method 1 0.883 1.697 our proposed method 2 0.727 1.566 smooth L1 + L1 center loss + sampling [3] 0.991 1.720 OSVR-L1 [21] 1.309 2.758 OSVR-L2 [21] 1.299 2.719 Table 2. Performance of our network when evaluated using the weighted MAE and weighted MSE proposed by [3].…”

Section: Discussionmentioning

confidence: 99%

Saliency Supervision: An Intuitive and Effective Approach for Pain Intensity Regression

Zhu

Zhao

2018

Neural Information Processing

View full text Add to dashboard Cite

show abstract

“…We only include the performance from previous studies who have reported their results on a 16-point scale for a fair comparison. Some previous studies [33,45] have adopted a simpler approach of reducing the original PSPI 16-point problem to a 6-point scale. The original 16 pain levels are discretized in a data balancing manner.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…Florea et al [12] proposed a histogram of topographical features for pain estimation in a transfer learning framework. Exploiting the temporal progression of pain expression from neutral through the apex and then back to neutral, Zhong et al [45] propose ordinal information for pain estimation.…”

Section: Automatic Pain Recognitionmentioning

confidence: 99%

“…Temporal information is encoded within the deep learned features and they show that this complementary information improves the recognition model. To deal with the problem of sparse data, previous works [33,45] have resorted to clustering the pain levels for better performance i.e. the 16-point PSPI scale is reduced to a 5-point problem where levels 4-5 are merged while levels 6+ are grouped into one class.…”

Section: Automatic Pain Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

Cumulative attributes for pain intensity estimation

Egede

Valstar

2017

Proceedings of the 19th ACM International Conference on Multimodal Interaction

View full text Add to dashboard Cite

Pain estimation from face video is a hard problem in automatic behaviour understanding. One major obstacle is the difficulty of collecting sufficient amounts of data, with balanced amounts of data for all pain intensity levels. To overcome this, we propose to adopt Cumulative Attributes, which assume that attributes for high pain levels with few examples are a superset of all attributes of lower pain levels. Experimental results show a consistent relative performance increase in the order of 20% regardless of features used. Our final system significantly outperforms the state of the art on the UNBC McMaster Shoulder Pain database by using cumulative attributes with Relevance Vector Regression on a combination of features, including appearance, geometric, and deep learned features.

show abstract