This paper describes an improved Continuous Dynamic Programming (CDP) for human gesture recognition. In order to cope with the variation in the amplitude of the input signal, a reference pattern set containing patterns with different mean amplitude is used instead of a single reference pattern. Next, a new multiple search path determination strategy is developed to replace the traditional single one to deal with the variation in the speed of the input gesture. In addition, a matching error controlled path determination method is newly introduced to reduce the influence of noise. The experimental results show this CDP can give stable recognition results compared with the conventional one.