Improving Pain Recognition Through Better Utilisation of Temporal Information.

摘要: Automatically recognizing pain from video is a very useful application as it has the potential to alert carers patients that are in discomfort who would otherwise not be able communicate such emotion (i.e young children, postoperative care etc.). In previous work [1], “pain-no pain” system was developed which used an AAM-SVM approach good effect. However, with any task involving large amount of data, there memory constraints need adhered and this compressing temporal signal using K-means clustering training phase. visual speech recognition, well known dynamics play vital role recognition. As recognition similar (i.e. recognising facial actions), our belief reduces likelihood accurately pain. paper, we show by spatial instead signal, achieve better Our results importance pain, however, do highlight some problems associated doing due randomness patient’s actions. Index Terms: expression, action units (AUs), active appearance models (AAM)