# Training Transformer Models by Wavelet Losses Improves Quantitative and Visual Performance in Single Image Super-Resolution arXiv: https://arxiv.org/abs/2404.11273 · HTML: https://mlsanity.com/paper/2404.11273/ ## Key facts - Cited by 10 papers in the map (data through 14 Oct 2026). - Builds on 41 of its 63 references that are in the map. - First posted on arXiv 17 Apr 2024. ## Builds on (most cited first, 30 of 41) - [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017) - [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021) - [Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions](https://mlsanity.com/paper/2102.12122/) (2021) - [SwinIR: Image Restoration Using Swin Transformer](https://mlsanity.com/paper/2108.10257/) (2021) - [Enhanced Deep Residual Networks for Single Image Super-Resolution](https://mlsanity.com/paper/1707.02921/) (2017) - [Image Super-Resolution Using Very Deep Residual Channel Attention Networks](https://mlsanity.com/paper/1807.02758/) (2018) - [ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks](https://mlsanity.com/paper/1809.00219/) (2018) - [CvT: Introducing Convolutions to Vision Transformers](https://mlsanity.com/paper/2103.15808/) (2021) - [Accurate Image Super-Resolution Using Very Deep Convolutional Networks](https://mlsanity.com/paper/1511.04587/) (2015) - [Pre-Trained Image Processing Transformer](https://mlsanity.com/paper/2012.00364/) (2020) - [Uformer: A General U-Shaped Transformer for Image Restoration](https://mlsanity.com/paper/2106.03106/) (2021) - [Residual Dense Network for Image Super-Resolution](https://mlsanity.com/paper/1802.08797/) (2018) - [Twins: Revisiting the Design of Spatial Attention in Vision Transformers](https://mlsanity.com/paper/2104.13840/) (2021) - [CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows](https://mlsanity.com/paper/2107.00652/) (2021) - [Activating More Pixels in Image Super-Resolution Transformer](https://mlsanity.com/paper/2205.04437/) (2022) - [Conditional Positional Encodings for Vision Transformers](https://mlsanity.com/paper/2102.10882/) (2021) - [Early Convolutions Help Transformers See Better](https://mlsanity.com/paper/2106.14881/) (2021) - [Single Image Super-Resolution via a Holistic Attention Network](https://mlsanity.com/paper/2008.08767/) (2020) - [Incorporating Convolution Designs into Visual Transformers](https://mlsanity.com/paper/2103.11816/) (2021) - [LocalViT: Analyzing Locality in Vision Transformers](https://mlsanity.com/paper/2104.05707/) (2021) - [Focal Frequency Loss for Image Reconstruction and Synthesis](https://mlsanity.com/paper/2012.12821/) (2020) - [Learning A Sparse Transformer Network for Effective Image Deraining](https://mlsanity.com/paper/2303.11950/) (2023) - [Efficient Long-Range Attention Network for Image Super-resolution](https://mlsanity.com/paper/2203.06697/) (2022) - [Dual Aggregation Transformer for Image Super-Resolution](https://mlsanity.com/paper/2308.03364/) (2023) - [Video Super-Resolution Transformer](https://mlsanity.com/paper/2106.06847/) (2021) - [Cross Aggregation Transformer for Image Restoration](https://mlsanity.com/paper/2211.13654/) (2022) - [Accurate Image Restoration with Attention Retractable Transformer](https://mlsanity.com/paper/2210.01427/) (2022) - [Fourier Space Losses for Efficient Perceptual Image Super-Resolution](https://mlsanity.com/paper/2106.00783/) (2021) - [SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution](https://mlsanity.com/paper/2208.11247/) (2022) - [NTIRE 2024 Challenge on Image Super-Resolution (x4): Methods and Results](https://mlsanity.com/paper/2404.09790/) (2024) ## Built on by (most cited first, 10 of 10) - [The Tenth NTIRE 2025 Image Denoising Challenge Report](https://mlsanity.com/paper/2504.12276/) (2025) - [NTIRE 2025 Challenge on Image Super-Resolution (x4): Methods and Results](https://mlsanity.com/paper/2504.14582/) (2025) - [The Fourth Challenge on Image Super-Resolution (×4) at NTIRE 2026: Benchmark Results and Method Overview](https://mlsanity.com/paper/2604.14558/) (2026) - [DWTGS: Rethinking Frequency Regularization for Sparse-view 3D Gaussian Splatting](https://mlsanity.com/paper/2507.15690/) (2025) - [Complex Wavelet Mutual Information Loss: A Multi-Scale Loss Function for Semantic Segmentation](https://mlsanity.com/paper/2502.00563/) (2025) - [AdaptSR: Low-Rank Adaptation for Efficient and Scalable Real-World Super-Resolution](https://mlsanity.com/paper/2503.07748/) (2025) - [GDSR: Global-Detail Integration through Dual-Branch Network with Wavelet Losses for Remote Sensing Image Super-Resolution](https://mlsanity.com/paper/2501.01460/) (2024) - [DiMoSR: Feature Modulation via Multi-Branch Dilated Convolutions for Efficient Image Super-Resolution](https://mlsanity.com/paper/2505.21262/) (2025) - [Wavelet-SARIMA-Transformer: A Hybrid Model for Rainfall Forecasting](https://mlsanity.com/paper/2509.11903/) (2025) - [UCAN: Unified Convolutional Attention Network for Expansive Receptive Fields in Lightweight Super-Resolution](https://mlsanity.com/paper/2603.11680/) (2026) ## Abstract (arXiv) Transformer-based models have achieved remarkable results in low-level vision tasks including image super-resolution (SR). However, early Transformer-based approaches that rely on self-attention within non-overlapping windows encounter challenges in acquiring global information. To activate more input pixels globally, hybrid attention models have been proposed. Moreover, training by solely minimizing pixel-wise RGB losses, such as L1, have been found inadequate for capturing essential high-frequency details. This paper presents two contributions: i) We introduce convolutional non-local sparse attention (NLSA) blocks to extend the hybrid transformer architecture in order to further enhance its receptive field. ii) We employ wavelet losses to train Transformer models to improve quantitative and subjective performance. While wavelet losses have been explored previously, showing their power in training Transformer-based SR models is novel. Our experimental results demonstrate that the proposed model provides state-of-the-art PSNR results as well as superior visual performance across various benchmark datasets.