Video Encoding Motion Search Using Depth Information
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding methods do not efficiently utilize depth information, leading to increased encoding complexity and reduced performance when encoding two-dimensional videos.
Innovation Solution
The method determines a motion searching range based on object information from depth images to calculate and optimize motion vectors for current coding units, reducing unnecessary calculations and improving encoding efficiency by restricting the search range to areas with correlated object regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a full motion searching region is used to calculate motion vectors, then the accuracy of motion vector determination is improved, but the encoding complexity and time consumption increase
Solution Approach 1:
The motion searching region is segmented into multiple sub-regions based on depth information. Instead of searching the entire region, the search is divided into smaller sub-regions corresponding to different depth ranges, allowing for more efficient and accurate motion vector determination while reducing overall computational complexity
Solution Approach 2:
Different search strategies are applied to different regions based on their depth characteristics. Regions with similar depth values use restricted search ranges, while regions with varying depth values use broader search ranges, optimizing the balance between accuracy and computational effort for each local area
2Reliability
If a full motion searching region is used to calculate motion vectors, then the probability of finding the optimum motion vector is improved, but the encoding time increases
Solution Approach 1:
Depth information is processed in advance to pre-classify pixels into different depth-based groups. This preliminary classification allows the motion search to start from already-organized sub-regions, reducing the time required to find optimal motion vectors while maintaining high probability of success
Solution Approach 2:
The motion searching range is dynamically adjusted based on depth information. For each coding unit, the search range is adaptively restricted to regions with similar depth characteristics, making the encoding process faster while maintaining reliability in finding optimal motion vectors
3Device complexity
If depth information is not utilized in video encoding, then the encoding process is simpler, but the encoding efficiency is reduced
Solution Approach 1:
The depth information, originally intended for 3D video applications, is repurposed for 2D video encoding to improve motion compensation efficiency. This multi-functional use of depth information enhances encoding productivity without significantly increasing process complexity
Solution Approach 2:
The encoding process incorporates depth information as an additional parameter to dynamically adjust motion searching ranges. This parameter change enables more efficient encoding by restricting searches to relevant regions, improving productivity while adding manageable complexity
Data Source
AI summary
An image encoding method using a depth information includes determining a motion searching range in a total motion searching region for calculating a motion vector of a current coding unit (CU) based on an object information of the current CU; and determining an optimum motion vector of the current CU based on a motion information inside the motion searching range.