Methods, apparatuses, and media for visual data processing
By dividing the residual sample set into subsets of multiple regions and introducing indication information into the bitstream, the problem of insufficient encoding and decoding efficiency in the existing technology is solved, and more efficient encoding and decoding processing is achieved.
CN122423201APending Publication Date: 2026-07-17DOUYIN CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DOUYIN CO LTD
- Filing Date
- 2024-12-18
- Publication Date
- 2026-07-17
Smart Images
Figure CN122423201A_ABST
Abstract
本公开的实施例提供了一种用于可视数据处理的解决方案。提出了一种用于可视数据处理的方法。该方法包括：利用基于神经网络（NN）的模型执行可视数据与可视数据的码流之间的转换，其中与可视数据相关联的残差样点集合被分割为与多个区域对应的残差样点的多个子集，并且码流包括第一指示，该第一指示用于指示针对多个区域中的每个区域的残差样点是否在码流的与该区域对应的子流中。
Need to check novelty before this filing date? Find Prior Art