The highly error-prone channel and limited computational power of terminal devices necessitates the implementation of robust yet simple error concealment. Error concealment techniques commonly make use of the surrounding correctly received image data or motion information for concealment. In this paper, we propose an efficient spatio-temporal boundary matching algorithm (ESTBMA) which exploits both spatial and temporal information to reconstruct the lost motion vectors (MV) and also introduces a new side smoothness measurement. The motion vector corresponding to the minimum of the distortion function is used as the estimation of motion vector of the lost block. Compared to the classical average motion vector (AVMV), median motion video (MMV) and boundary matching algorithm (BMA), simulation results show that the proposed algorithm can recover the higher quality image on both subjective visual evaluation and objective numerical metrics.