关键词: 手势识别, 深度数据, 深度摄像头, 手势数据集

Abstract: Hand gesture is one of the most important means in human-computer interaction. Due to the characteristics of non-contact interaction，vision-based hand gesture recognition has been more and more widely used in the area of artificial intelligence. However，due to the restriction of the conventional 2D optical camera，the quality of collected gesture images is easily affected by the light conditions and cluttered backgrounds，which is a major challenge in hand gesture extraction，and severely restricts the practical application of vision-based hand gesture recognition. In recent years，the emergence of depth camera technologies has brought new opportunities to address the above problems. With the help of depth data，new vision-based hand gesture recognition approaches are emerging in an endless stream， and the recognition accuracy is steadily improved，which effectively promotes the practical application of vision-based hand gesture recognition.In this context，from the data angle，we reviewed the latest developments of the depth data based hand gesture recognition from three aspects：depth data acquisition，common hand gesture datasets and depth data based hand gesture recognition methods，and make forecasts on its future developments.

Key words: hand gesture recognition, depth data, depth camera, hand gesture dataset