SMART network advances light field super-resolution with novel Mamba and Transformer integration

Researchers have developed a new hybrid network called SMART for Light Field Super-Resolution (LFSR). This model integrates a Slope-Guided Mamba and an Angular-Refined Transformer to better capture spatial-angular correlations and maintain 4D ray coherence. SMART addresses limitations in existing methods by bridging the gap between spatial and angular dimensions and enabling geometry-consistent sequence modeling along epipolar structures. Experiments show SMART achieves state-of-the-art performance, outperforming previous methods by 0.42 dB in PSNR and reducing artifacts. AI

IMPACT This novel network architecture could advance image processing capabilities in fields requiring high-fidelity spatial-angular data.

RANK_REASON The item is a research paper detailing a new technical approach to a specific computer vision problem. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• alphaXiv
• Angular-Refined Transformer
• arXiv
• CatalyzeX Code Finder for Papers
• Connected Papers
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations
• Slope-Guided Mamba
• SMART

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →