AI RESEARCH
SAT: Selective Aggregation Transformer for Image Super-Resolution
arXiv CS.CV
•
ArXi:2604.07994v1 Announce Type: new Transformer-based approaches have revolutionized image super-resolution by modeling long-range dependencies. However, the quadratic computational complexity of vanilla self-attention mechanisms poses significant challenges, often leading to compromises between efficiency and global context exploitation. Recent window-based attention methods mitigate this by localizing computations, but they often yield restricted receptive fields. To mitigate these limitations, we propose Selective Aggregation Transformer