Remote sensing images often suffer from cloud cover. Cloud removal is required in many applications of remote sensing images. Multitemporal-based methods are popular and effective to cope with thick clouds. This paper contributes to a summarization and experimental comparation of the existing multitemporal-based methods. Furthermore, we propose a spatiotemporal-fusion with poisson-adjustment method to fuse multi-sensor and multi-temporal images for cloud removal. The experimental results show that the proposed method has potential to address the problem of accuracy reduction of cloud removal in multi-temporal images with significant changes.