Object-Based Intra-Prediction for Video Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding and decoding techniques are inefficient in predicting pixel values for blocks with multiple directional features, as they rely on directional similarities which do not effectively capture the redundancy in such blocks.
Innovation Solution
Implementing object-based intra-prediction methods that segment blocks into portions based on defined shapes and use specific prediction modes for each portion, combining object-based and directional prediction modes to efficiently encode and decode video signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If directional prediction modes are used for blocks with multiple directional features, then the encoding process is simple, but the prediction accuracy is insufficient
Solution Approach 1:
The current block is divided into multiple spatial portions based on detected edges and lines. Each portion is then encoded using a prediction mode suited to its characteristics, allowing different directional features to be captured accurately while maintaining manageable encoding complexity through systematic segmentation.
Solution Approach 2:
Different prediction modes are applied to different spatial portions of the block based on local features. Edge portions use edge-directed prediction, line portions use line-directed prediction, and other portions use standard directional prediction, optimizing prediction accuracy for each local region while keeping the overall process structured.
2Measurement precision
If object-based intra-prediction with multiple prediction modes is implemented, then prediction accuracy for blocks with multiple directional features is improved, but the device complexity increases
Solution Approach 1:
The block is segmented into spatial portions based on detected edges and lines, with each portion assigned an appropriate prediction mode. This segmentation approach improves prediction accuracy by matching prediction strategies to local features while organizing the complexity into manageable, systematic steps.
Solution Approach 2:
The prediction mode selection is dynamic rather than static, allowing the encoder to adaptively choose different prediction modes for different spatial portions based on detected features. This dynamic adaptation improves accuracy while the systematic decision process keeps complexity manageable.
3Productivity
If the current block is segmented into multiple spatial portions with different prediction modes, then compression efficiency is improved, but the processing time increases
Solution Approach 1:
The block is divided into spatial portions based on detected edges and lines, allowing each portion to be encoded with an optimized prediction mode. This segmentation improves compression efficiency by better matching the encoding strategy to local features, while the systematic approach to segmentation and encoding keeps processing time manageable.
Solution Approach 2:
Edge and line detection is performed as a preliminary step before segmentation and encoding. This preliminary analysis identifies the optimal segmentation strategy in advance, allowing the encoder to efficiently process each spatial portion with the appropriate prediction mode, improving overall compression efficiency while organizing processing time into distinct, manageable phases.
Data Source
AI summary
Object-based intra-prediction encoding may include generating, by a processor in response to instructions stored on a non-transitory computer readable medium, an encoded block of a current frame of a video stream by encoding a current block from the current frame, including the encoded block in an output bitstream, and outputting or storing the output bitstream. Encoding the current block may include identifying a first spatial portion of the current block, wherein the first spatial portion includes a first pixel from the current block and omits a second pixel from the current block, encoding the first pixel using a first intra-prediction mode, and encoding the second pixel using a second intra-prediction mode, wherein the second intra-prediction mode differs from the first intra-prediction mode.