The accuracy of depth scene from Kinect will always be influenced by m any surrounding factors.Therefore missing areas will appear in the depth map obtained by depth camera.This paper proposes an inpainting algorithm based on superpixel segmentation to improve the quality of depth map.Firstly,the depth map and color map will be filtered and segmented,respect ively.Secondly, the similarity of color image and depth map will be used for K-means cluster,an d each cluster position is recorded as labels which will be mapped to the depth map.Finally,c lusters are divided into four types according to the ratio of the missing pixels in each clu ster.The types include cluster without holes,cluster with small holes,cluster with large hole s and all-hole cluster.The clusters with small holes and large holes will be filled with fast m arching method and median filling algorithm,respectively.Besides,we use nearly similar area f illing method for the all-hole cluster,and the processing of cluster without holes is unneces sary.Experimental results show that the residual mean square error of the proposed method is lower than that of FMM,Shen′s and Scheming′s methods by 2.9584,0.82249and 0.07840,respectively.Besides,the subjective quality of t he proposed method is better than that of those methods.