In video-assisted scenarios such as live events, remote surgery, autonomous driving and so on, age of information (AoI) can be used as a quality of service (QoS) metric for evaluating video transmission quality in conjunction with effective capacity (EC) due to its unique property of portraying the freshness of information. In this paper, we propose a Multiple-Input Multiple-Output (MIMO) downlink video transmission system and apply the AoI model of the idealized service process to the video transmission system to obtain an upper bound on the peak AoI violation probability. Considering the limited resource and the information freshness requirements of video users simultaneously, we formulate the total effective capacity maximization problem. By decomposing the original problem into two resource allocation subproblems, an iterative resource allocation algorithm is proposed to find the optimal solution. The simulation results show that the proposed algorithm has good convergence performance and can obtain the corresponding optimal EC values for different numbers of users. Meanwhile, the optimal EC value in the proposed algorithm is greater than that in the two baselines.