An Image-Content-Based Adaptive Tile Partitioning Algorithm

In this paper, an adaptive segmentation algorithm based on image content is proposed, which aims to reduce the encoding loss generated by the process of utilizing the new feature tile in the video coding standard. The algorithm analyzes the regional correlation of the current frame before the encoder performs the current frame encoding, so as to determine the critical area of the correlation change in the frame, divide the tile in the critical area, and ensure that the area with high correlation is not divided by the tile as much as possible. Experimental results show that compared with the case of uniform division of Tile, the proposed algorithm has improved performance.