Multi-Attention Based Ultra Lightweight Image Super-Resolution
arXiv:2008.12912
Abstract
Lightweight image super-resolution (SR) networks have the utmost significance for real-world applications. There are several deep learning based SR methods with remarkable performance, but their memory and computational cost are hindrances in practical usage. To tackle this problem, we propose a Multi-Attentive Feature Fusion Super-Resolution Network (MAFFSRN). MAFFSRN consists of proposed feature fusion groups (FFGs) that serve as a feature extraction block. Each FFG contains a stack of proposed multi-attention blocks (MAB) that are combined in a novel feature fusion structure. Further, the MAB with a cost-efficient attention mechanism (CEA) helps us to refine and extract the features using multiple attention mechanisms. The comprehensive experiments show the superiority of our model over the existing state-of-the-art. We participated in AIM 2020 efficient SR challenge with our MAFFSRN model and won 1st, 3rd, and 4th places in memory usage, floating-point operations (FLOPs) and number of parameters, respectively.
ECCVW AIM2020
References in corpus (9)
- Adam: A Method for Stochastic Optimization
- Distilling the Knowledge in a Neural Network
- ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks
- Do Deep Convolutional Nets Really Need to be Deep and Convolutional?
- Fast, Accurate and Lightweight Super-Resolution with Neural Architecture Search
- Lightweight Image Super-Resolution with Adaptive Weighted Learning Network
- AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights
- Hybrid Residual Attention Network for Single Image Super Resolution
- MAMNet: Multi-path Adaptive Modulation Network for Image Super-Resolution