Enhanced Template-based Intra Mode Derivation with Adaptive Block Vector Replacement

2025-09-29

AI Summary

Conventional intra prediction relies solely on neighboring pixels for mode decision, overlooking potential spatial similarities in non-adjacent regions and thus limiting prediction efficiency. To address this, we propose a template-matching-based intra mode derivation enhancement method that innovatively introduces adaptive block-vector replacement and extended reference range mechanisms—marking the first integration of non-adjacent region information into intra mode decision. Crucially, the method preserves the standard coding framework and maintains identical encoder/decoder complexity. Experimental results under all-intra configuration demonstrate an average 0.082% BD-rate reduction for the Y component compared to ECM-16.1; screen content sequences achieve an additional 0.25% gain. These results validate both the effectiveness and practicality of modeling non-adjacent spatial similarity for intra prediction.

Technology Category

• Computer Vision: Multi-modal Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Intra prediction is a crucial component in traditional video coding frameworks, aiming to eliminate spatial redundancy within frames. In recent years, an increasing number of decoder-side adaptive mode derivation methods have been adopted into Enhanced Compression Model (ECM). However, these methods predominantly rely on adjacent spatial information for intra mode decision-making, overlooking potential similarity patterns in non-adjacent spatial regions, thereby limiting intra prediction efficiency. To address this limitation, this paper proposes a template-based intra mode derivation approach enhanced by block vector-based prediction. The adaptive block vector replacement strategy effectively expands the reference scope of the existing template-based intra mode derivation mode to non-adjacent spatial information, thereby enhancing prediction efficiency. Extensive experiments demonstrate that our strategy achieves 0.082% Bjøntegaard delta rate (BD-rate) savings for Y components under the All Intra (AI) configuration compared to ECM-16.1 while maintaining identical encoding/decoding complexity, and delivers an additional 0.25% BD-rate savings for Y components on screen content sequences.

Problem

• Improves intra prediction using non-adjacent spatial information
• Expands template-based mode derivation with adaptive block vectors
• Enhances coding efficiency while maintaining identical complexity

Innovation

• Template-based intra mode derivation with block vector prediction
• Adaptive block vector replacement expands reference scope
• Enhanced prediction efficiency using non-adjacent spatial information