Аңдатпа
This article is aimed at illustrating the approach of detecting and recognizing Kazakh Sign Language static gestures without using expensive depth cameras or sensor gloves. For this research a simple RGB mono camera had been used. The main distinguishing feature of our approach is that, unlike most of the existing approaches, it performs hand detection and gesture recognition jointly, i.e. at the very moment when the shape of a static gesture is detected its specific label is automatically recognized. The approach was tested on a range of datasets, and encouraging results were achieved.