Image encoding process, image decoding process method, electronic device, and program product
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA MOBILE COMM GRP SHAANXI CO LTD
- Filing Date
- 2026-04-20
- Publication Date
- 2026-07-24
AI Technical Summary
Existing technologies do not fully consider the differences in importance between regions of interest (ROIs) and non-ROIs in video content, resulting in poor video compression performance, making it difficult to meet the requirements of real-time performance and low storage costs. Furthermore, existing methods are easily affected by background interference and target diversity when identifying ROIs, leading to inaccurate identification results.
By extracting partitioned feature maps of key image data through a recurrent memory network, and combining adaptive dynamic offset to correct the position parameters of the region of interest, global and local attention weights are determined, and weighted calculation and encoding are performed to form an end-to-end image coding processing method.
It improves the accuracy of region of interest identification and coding efficiency, reduces the feature weights of non-regions of interest, lowers computational complexity and time latency, meets the requirements of real-time performance and low storage cost, and improves the quality and efficiency of video compression.
Smart Images

Figure CN122457777A_ABST