The Joint Video Experts Team (JVET) is exploring next-generation video coding technologies through the Enhanced Compression Model (ECM). In ECM-16.1, the Non-Cross-Component Prediction (Non-CCP) chroma candidate list combines six initial candidates with angular and Direct Block Vector (DBV) candidates derived from predefined positions. However, the existing derivation process considers only limited positions and does not exploit the Luma Block Vector (BV) history buffer, resulting in few additional distinct candidates. Moreover, early termination may stop candidate evaluation before potentially useful candidates are sufficiently examined. This paper extends the derivation positions for angular and DBV candidates, reuses BVs from previously coded DBV blocks at neighboring chroma positions, and exploits the luma BV history buffer for additional DBV candidate derivation. Candidate evaluation is further refined by relaxing the early termination condition and applying the sum of absolute differences to both luma and chroma template costs. Under the All Intra configuration of the JVET common test conditions, average Bjøntegaard Delta rate changes of −0.04%, −0.08%, and −0.01% are achieved for Y, Cb, and Cr, respectively, compared with ECM-16.1. For screen content, the corresponding changes are −0.12%, −0.25%, and −0.18%, with encoder and decoder runtime ratios of 100.4% and 100.9%.
No takes yet. Share an insight, caveat, or question.
Lee et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: