Perceive the difference!
The recently developed video coding standard, H.264 / AVC, significantly outperforms previous standards in terms of coding performance at reasonable implementation complexity. Several application systems, such as high-definition DVD and digital video broadcasting for handheld devices and high-definition television systems, have adopted H.264 or its modified versions as the video coding standard. In addition, the extensions of H.264 / AVC to scalable and multi-view video coding applications are nearly finalized. Many video services, especially bandwidth-limited wireless video, will benefit from the H.264 coder due to its outstanding features. The use of variable block sizes for intra- and inter-prediction in combination with different prediction modes and motion compensation using multiple reference frames is one of the main reasons for the improved coding efficiency in H.264 / AVC. Together with many other new features, the encoder can select between multitudes of different coding modes. The determination of the optimal coding mode under the joint rate and distortion consideration, which is called the rate-distortion optimization (RDO), introduces a huge amount of memory access and computational complexity for testing all possible modes in video encoders. Hence, a reduction of the complexity of motion estimation and mode selection in an H.264/AVC encoder becomes an important task for real-time applications.