Multicodec processing and rate control
By using multiple codec selection units and optimizing the cost function, combined with neural networks, the signal is decomposed into multiple components and encoded separately, solving the problem of low compression efficiency in existing technologies and achieving the best encoding quality and efficiency for video sequences.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- V NOVA INT LTD
- Filing Date
- 2019-01-17
- Publication Date
- 2026-07-10
AI Technical Summary
Existing compression technologies fail to effectively utilize the characteristics of different compression techniques when encoding video sequences, resulting in low encoding efficiency and an inability to achieve optimal encoding quality for each part of the video sequence.
The signal is decomposed into multiple components by using a multi-coder selection unit, and each component is encoded using a different encoding algorithm. The bit rate allocation of each component is determined by optimizing the cost function, and the encoding process is optimized by combining neural networks or deep learning algorithms.
It enables the use of the best combination of compression techniques in each part of the video sequence, optimizing encoding performance and quality while maintaining reverse compatibility, thus improving encoding efficiency and quality.
Smart Images
Figure CN111869206B_ABST