作者: Cheolhong An , Liang Liang , Szepo Robert Hung
DOI:
关键词: Motion compensation 、 Video tracking 、 Computer vision 、 Encoder 、 Video camera 、 Quantization (image processing) 、 Artificial intelligence 、 Reference frame 、 Motion estimation 、 Computer science 、 ENCODE
摘要: This disclosure describes techniques for improving functionalities of a back-end device, e.g., video encoder, using parameters detected and estimated by front-end camera. The may involve estimating blurriness level associated with frames captured during refocusing process. Based on the level, quantization parameter (QP) used to encode blurry is adjusted either in camera or encoder. encoder uses QP frames. also estimate adjust encoding algorithms simplifying motion estimation compensation