Fast Block Direction Prediction for Directional Transforms

DCT-based transforms are widely adopted for video compression. Recently, many authors have highlighted that prediction residuals usually have directional structures that cannot be efficiently represented by conventional Discrete Cosine Transform (DCT). Although directional transforms have superior performance over the conventional DCT, for application in video compression it is necessary to evaluate increase in coding time and complexity for its implementation. This paper proposes a fast algorithm for estimating blocks directions before applying directional transforms. The encoder identifies predominant directions in each block, and only applies the transform referent to that direction. The algorithm can be used in conjunction with any proposed algorithm for directional transforms that uses the RDO process for selection of the direction to be explored, reducing implementation complexity to similar levels when only conventional DCT is used. For the tested proposal, estimated reduction in total coding time would be 37.5% (4 × 4 blocks with 8 possible directions) and 47.4% (8× 8 blocks with 16 possible directions).