# MFAGAN: A Compression Framework for Memory-Efficient On-Device Super-Resolution GAN

arXiv: https://arxiv.org/abs/2107.12679 · HTML: https://mlsanity.com/paper/2107.12679/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 43 references that are in the map.
- First posted on arXiv 27 Jul 2021.

## Builds on (most cited first, 30 of 33)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Perceptual Losses for Real-Time Style Transfer and Super-Resolution](https://mlsanity.com/paper/1603.08155/) (2016)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network](https://mlsanity.com/paper/1609.04802/) (2016)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [Neural Architecture Search with Reinforcement Learning](https://mlsanity.com/paper/1611.01578/) (2016)
- [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)
- [Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network](https://mlsanity.com/paper/1609.05158/) (2016)
- [Residual Dense Network for Image Super-Resolution](https://mlsanity.com/paper/1802.08797/) (2018)
- [Once-for-All: Train One Network and Specialize it for Efficient Deployment](https://mlsanity.com/paper/1908.09791/) (2019)
- [FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search](https://mlsanity.com/paper/1812.03443/) (2018)
- [Accelerating the Super-Resolution Convolutional Neural Network](https://mlsanity.com/paper/1608.00367/) (2016)
- [Deeply-Recursive Convolutional Network for Image Super-Resolution](https://mlsanity.com/paper/1511.04491/) (2015)
- [EIE: Efficient Inference Engine on Compressed Deep Neural Network](https://mlsanity.com/paper/1602.01528/) (2016)
- [Fast, Accurate, and Lightweight Super-Resolution with Cascading Residual Network](https://mlsanity.com/paper/1803.08664/) (2018)
- [Lightweight Image Super-Resolution with Information Multi-distillation Network](https://mlsanity.com/paper/1909.11856/) (2019)
- [EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis](https://mlsanity.com/paper/1612.07919/) (2016)
- [An Energy and GPU-Computation Efficient Backbone Network for Real-Time Object Detection](https://mlsanity.com/paper/1904.09730/) (2019)
- [RankSRGAN: Generative Adversarial Networks with Ranker for Image Super-Resolution](https://mlsanity.com/paper/1908.06382/) (2019)
- [Residual Feature Distillation Network for Lightweight Image Super-Resolution](https://mlsanity.com/paper/2009.11551/) (2020)
- [Fast and Accurate Single Image Super-Resolution via Information Distillation Network](https://mlsanity.com/paper/1803.09454/) (2018)
- [AutoGAN: Neural Architecture Search for Generative Adversarial Networks](https://mlsanity.com/paper/1908.03835/) (2019)
- [Patch-Based Image Inpainting with Generative Adversarial Networks](https://mlsanity.com/paper/1803.07422/) (2018)
- [Fast, Accurate and Lightweight Super-Resolution with Neural Architecture Search](https://mlsanity.com/paper/1901.07261/) (2019)
- [Densely Residual Laplacian Super-Resolution](https://mlsanity.com/paper/1906.12021/) (2019)
- [AutoGAN-Distiller: Searching to Compress Generative Adversarial Networks](https://mlsanity.com/paper/2006.08198/) (2020)
- [GAN Slimming: All-in-One GAN Compression by A Unified Optimization Framework](https://mlsanity.com/paper/2008.11062/) (2020)
- [Distilling portable Generative Adversarial Networks for Image Translation](https://mlsanity.com/paper/2003.03519/) (2020)
- [AIM 2020 Challenge on Efficient Super-Resolution: Methods and Results](https://mlsanity.com/paper/2009.06943/) (2020)

## Built on by (most cited first, 14 of 14)

- [SwinIR: Image Restoration Using Swin Transformer](https://mlsanity.com/paper/2108.10257/) (2021)
- [Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling](https://mlsanity.com/paper/2108.05301/) (2021)
- [CiaoSR: Continuous Implicit Attention-in-Attention Network for Arbitrary-Scale Image Super-Resolution](https://mlsanity.com/paper/2212.04362/) (2022)
- [Mutual Affine Network for Spatially Variant Kernel Estimation in Blind Image Super-Resolution](https://mlsanity.com/paper/2108.05302/) (2021)
- [Generative Adversarial Networks for Image Super-Resolution: A Survey](https://mlsanity.com/paper/2204.13620/) (2022)
- [A Dive into SAM Prior in Image Restoration](https://mlsanity.com/paper/2305.13620/) (2023)
- [Reconstructing High-resolution Turbulent Flows Using Physics-Guided Neural Networks](https://mlsanity.com/paper/2109.03327/) (2021)
- [Efficient Image Deblurring Networks based on Diffusion Models](https://mlsanity.com/paper/2401.05907/) (2024)
- [AnySR: Realizing Image Super-Resolution as Any-Scale, Any-Resource](https://mlsanity.com/paper/2407.04241/) (2024)
- [Design Automation for Fast, Lightweight, and Effective Deep Learning Models: A Survey](https://mlsanity.com/paper/2208.10498/) (2022)
- [Physics-enhanced Neural Operator for Simulating Turbulent Transport](https://mlsanity.com/paper/2406.04367/) (2024)
- [Reconstructing Turbulent Flows Using Physics-Aware Spatio-Temporal Dynamics and Test-Time Refinement](https://mlsanity.com/paper/2304.12130/) (2023)
- [Modeling Continuous Spatial-temporal Dynamics of Turbulent Flow with Test-time Refinement](https://mlsanity.com/paper/2412.19927/) (2024)
- [BrainPuzzle: Hybrid Physics and Data-Driven Reconstruction for Transcranial Ultrasound Tomography](https://mlsanity.com/paper/2510.20029/) (2025)

## Abstract (arXiv)

Generative adversarial networks (GANs) have promoted remarkable advances in single-image super-resolution (SR) by recovering photo-realistic images. However, high memory consumption of GAN-based SR (usually generators) causes performance degradation and more energy consumption, hindering the deployment of GAN-based SR into resource-constricted mobile devices. In this paper, we propose a novel compression framework Multi-scale Feature Aggregation Net based GAN (MFAGAN) for reducing the memory access cost of the generator. First, to overcome the memory explosion of dense connections, we utilize a memory-efficient multi-scale feature aggregation net as the generator. Second, for faster and more stable training, our method introduces the PatchGAN discriminator. Third, to balance the student discriminator and the compressed generator, we distill both the generator and the discriminator. Finally, we perform a hardware-aware neural architecture search (NAS) to find a specialized SubGenerator for the target mobile phone. Benefiting from these improvements, the proposed MFAGAN achieves up to 8.3× memory saving and 42.9× computation reduction, with only minor visual quality degradation, compared with ESRGAN. Empirical studies also show ~70 milliseconds latency on Qualcomm Snapdragon 865 chipset.