It should be appreciated that the apparatus and/or the process 200 may operate on a compressed video. For example, the apparatus may receive a compressed video and then the content decoder 122 of FIG. 1 may decode frames of the compressed video. The apparatus may then process the decoded frames to facilitate enhancing the quality of the video. For example, the device 100, the processing unit 120 and/or the video enhancer 198 may utilize spatial information and temporal information for facilitating video quality enhancement. Accordingly, the techniques disclosed herein may be computationally efficient as they process one frame at a time (e.g., the previous frame is used for optical flow estimation and may be accessed via a frame buffer). Additionally, the feedback from the output may enable the techniques disclosed herein to have increased temporal memory about past frames compared to multi-frame approaches that have no temporal memory and/or limited temporal memory. For example, the process 200 may align the previously enhanced frame (Yt?1) 234 based on the optical flow 212 associated with the current decoded frame (Xt) 202 to enhance the current decoded frame (Xt) 202. The feedback also may enable an output frame sequence with improved consistency. For example, the enhancements applied to the previous frame may be used while processing the current frame. The reuse of information reduces the likelihood of flicker in the enhanced video.
