Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
June 17, 2026MathematicsOpen Access

Luma Background Restoration for Semantic Segmentation in Video Coding for Machines

View Full Paper
Ask AI
Bookmark
Share

Authors

SKSeon‐Jae KimTLTaesik LeeBPByeongju Park

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates improved segmentation accuracy in video coding for machines, indicating enhanced object detection and tracking capabilities.

Key Points

  • To improve segmentation accuracy by restoring degraded background regions in video coding for machine vision tasks.
  • Proposed a luma background restoration method using structural correlation between luma and chroma components.
  • Integrated multi-channel linear modeling with context-based arithmetic coding for efficient transmission of indices.
  • Conducted experiments under VCM test conditions to evaluate performance.
  • Achieved an average Bjøntegaard Delta mean Intersection-over-Union (BD-mIoU) of 7.70.
  • Outperformed the latest background preservation method, which achieved a BD-mIoU of 7.41.
  • Effectively restored structural background details necessary for semantic segmentation.

Cite This Study

Kim et al. (2026) studied this question.

synapsesocial.com/papers/6a323bbcd50b63ecad2063a5https://doi.org/10.3390/math14122124
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Adaptive Chroma Block Vector Derivation from Luma for Screen Content Coding2024 · 1 citations
  2. 2Advanced fine-tuning procedures to enhance DNN robustness in visual coding for machines2024
  3. 3Suboptimal video coding for machines method based on selective activation of in‐loop filter2024 · 1 citations
  4. 4Perceptual Video Coding for Machines via Satisfied Machine Ratio Modeling2024 · 8 citations
  5. 5Object Segmentation-Assisted Inter Prediction for Versatile Video Coding2024