The current supervised face forgery video detection methods need a large amount of labeled data. In order to solve the practical problems of fast iteration and many kinds of video forgery methods， the unsupervised idea in temporal anomaly detection was introduced into face forgery video detection， the face forgery video detection task was transformed into unsupervised video anomaly detection task， and an unsupervised face forgery video detection method based on reconstruction error was proposed. Firstly， the facial landmark sequence of continuous frames in the video to be detected was extracted. Secondly， the facial landmark sequence in the video to be detected was reconstructed based on multi-granularity information such as deviation features， local features and temporal features. Thirdly， the reconstruction error between the original sequence and the reconstructed sequence was calculated. Finally， the score was calculated according to the peak frequency of the reconstruction error to detect the forgery video automatically. Experimental results show that compared with detection methods such as LRNet （Landmark Recurrent Network） and Xception-c23， the proposed method has the AUC （Area Under Curve） of the detection performance increased by up to 27.6%， and the AUC of the transplantation performance increased by 30.4%.

The H.264 communication over network may cause a whole frame loss. To solve the problem, an error-concealment algorithm based on residual distribution was proposed for whole frame packet loss in H.264. Firstly, the residual information of reference frame was analyzed. Then, according to the result, the motion vector copy algorithm was used in the region where the image was smooth or the image object's motion was rigid. For the other regions, the motion vector of each pixel was re-estimated, and then optical flow algorithm was introduced. Experimental results show that the algorithm outperforms the traditional one on both visual quality and Peak Signal-to-Noise Ratio (PSNR).