Block-based video coding algorithms such as the JPEG and MPEG are widely used in video communication systems over lossy channels. However, because current video communication channels do not guarantee visual quality of service during video transmission, random bit errors or erasure errors in the transmission data often occur, resulting in error blocks. Therefore, an efficient error concealment technique is needed to reconstruct error blocks and to preserve the geometric information of the original image. In this paper, a geometric-based error concealment technique for concealing transmission errors and improving visual quality is proposed by using the support vector machine (SVM) and distributed genetic algorithm (DGA). The technique smoothly corrects transmission errors in geometric information. The proposed technique can be summarized as follows: first, the error block is replaced a block estimated by the SVM. Then the geometric information of the replaced error block is reconstructed by the DGA. Experimental results are presented to illustrate the effectiveness of the proposed method