Abstract: Aiming at the problem of high complexity of existing video encryption algorithms, a low complexity texture-adaptive video encryption algorithm is proposed. Firstly, the quantized residual DCT coefficients and motion vectors obtained in the video encoding process are used to build the threshold-adaptive texture estimation model, which is used to detect texture region for video encryption. Then, considering the video encoding efficiency and format compatibility, the signs of quantized non-zero DCT coefficient and non-zero MVD are selected for encryption in the texture region, respectively. The algorithm takes the H.264/AVC video codec as a verification example, the experimental results show that, the proposed encryption algorithm ensures the security of video content and the efficiency of compression encoding, while reducing the complexity of the encryption algorithm and the amount of data to be encrypted greatly. Meanwhile, compared with existing texture detection methods, the complexity of the proposed texture detection algorithm is extremely low, and it can be applied to real-time video encryption application scenarios.