Volatile memories like double data rate random access memory or non-volatile memories like embedded flash memory are key components in most of the embedded systems, especially in today’s mobile video processing systems. There is a requirement to use the on-chip and offchip embedded memories very efficiently to meet the end user requirements. Increasingly dominating power consumption and shortening battery life of mobile devices is a key factor which is to be considered while designing and developing the embedded systems and embedded software solutions. Traditional hardware-level power optimization techniques usually come with significant implementation overhead to solve the memory failure problems during lowvoltage operations. This paper presents advanced mobile video memory optimization techniques which indirectly reduces power consumption and increases the overall system performance. The viewing contexts significantly influence the quality of the viewer experience, and in the context with higher quality, mobile users have higher tolerance to the video degradation. Accordingly, the memory failures can be introduced adaptively to achieve power savings without influencing the viewer experience. To meet the silicon area constraint in mobile devices, a simple but an efficient hardware implementation scheme can be developed based on these optimization techniques to minimize area overhead. Input mp4 video will be processed by various techniques like bicubic, bilinear, nearest neighbor and H264 compression to generate output mp4 videos with various memory sizes. The simulation results will also be provided.