Motion Vector Re-estimation for Video Trans-coding with Arbitrary Downsizing Many visual devices are applied to mobile phone, PDA, or other products lately. These devices can receive real-time video from wireless network or internet. In order to watch video programs, the video frame size received from various devices must be changed. Although many fast down-scaling algorithms have been proposed, most of traditional schemes focus mainly on downsizing the video by an integral factor. They are not suitable to adapt arbitrary video downscaling. Based on the object-based assumption, we introduce a distance-trimmed filter (DTF) to improve accuracy of estimated motion vectors (MVs). Furthermore, we use bilinear interpolation to resolve the problem of down sampling of float point factor. Therefore, we can downsize primitive frame of video to arbitrary factor to suit the various size of monitor of digital products. Finally, we use Kalman filter to refine accuracy of the estimated MVs. The experimental results indicate that the performance of the proposed method for fast downsizing video provides an effective improvement.