Motion estimation (ME) has a vital role in video coding and several video processing applications, such as denoising, de-interlacing, and frame rate up-conversion (FRUC) or frame interpolation. ME is employed to exploit the temporal correlation between video frames either to reduce the temporal redundancy for video coding applications or to improve the visual video quality for video processing applications. One might argue that some of these video processing applications may potentially utilize the existing motion vectors (MVs) from the decoder via MV postprocessing to keep the complexity low; however, this may not usually be a feasible option. This infeasibility could be due to either difficulty of using MVs or lack of available MVs. As video coding and video processing applications are often implemented separate intellectual properties (IPs) in hardware[12], it may be very difficult to share the MVs between decoder and other video processing applications due to bandwidth, latency, storage, and design specification reasons. Besides, some of these video processing applications may be employed either before the encoding or after the decoding, and some of them may be employed at both places; if it is employed before the encoding then MVs are not available, as a result ME needs to be performed. For example, FRUC is employed only at the display side after the decoder; de-interlacing and de-noising, however, can be utilized in both places. Where as in true motion estimation the mainly it goes to detect the motion object as closely as possible by using the block matching algorithm, and then after the estimation of the true motion vector fields it helps to produce the motion compensated temporal frame interpolation. This methods is gives the more video quality and the smoothness with the flow of frames. The main aim of this paper is to determine the motion (moving) object in the video sequences this method is called as true motion estimation by adopting the implicit and explicit smoothness constraint on block matching algorithm. After finding true motion vector also called as coherent motion vector field is used to produce the good temporal interpolated frames between existing frames this gives good video with easily flowing one after the other by smoothly and continuously. After getting the interpolated frames the performance metrics like PSNR (peak signal to noise ratio) and SSIM (structural similarity) between the interpolated frames and the original frames.