mlsanityData to 7 Oct 2026

Extremely Lightweight Quantization Robust Real-Time Single-Image Super Resolution for Mobile Devices

arXiv 2105.10288 ↗First posted 21 May 2021cs.CV cs.AI cs.LGcited by 30

Not on the map. Why?

Abstract (arXiv)

Single-Image Super Resolution (SISR) is a classical computer vision problem and it has been studied for over decades. With the recent success of deep learning methods, recent work on SISR focuses solutions with deep learning methodologies and achieves state-of-the-art results. However most of the state-of-the-art SISR methods contain millions of parameters and layers, which limits their practical applications. In this paper, we propose a hardware (Synaptics Dolphin NPU) limitation aware, extremely lightweight quantization robust real-time super resolution network (XLSR). The proposed model's building block is inspired from root modules for Image classification. We successfully applied root modules to SISR problem, further more to make the model uint8 quantization robust we used Clipped ReLU at the last layer of the network and achieved great balance between reconstruction quality and runtime. Furthermore, although the proposed network contains 30x fewer parameters than VDSR its performance surpasses it on Div2K validation set. The network proved itself by winning Mobile AI 2021 Real-Time Single Image Super Resolution Challenge.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Builds on 29 of its 33 references that are in the map. First posted on arXiv 21 May 2021.

Builds on

29 of 33 references.

References · 29

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  2. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  3. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  4. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  5. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  6. 2018ESRGAN: Enhanced Super-Resolution Generative Adversarial Networkscited by 1,250
  7. 2016Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Networkcited by 1,199
  8. 2017ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devicescited by 1,194
  9. 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,097
  10. 2016SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model sizecited by 1,054
  11. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 902
  12. 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 697
17 more
  1. 2015Cyclical Learning Rates for Training Neural Networkscited by 627
  2. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 557
  3. 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 456
  4. 2016Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activationscited by 413
  5. 2016Accelerating the Super-Resolution Convolutional Neural Networkcited by 393
  6. 2020Single Image Super-Resolution via a Holistic Attention Networkcited by 265
  7. 2019Lightweight Image Super-Resolution with Information Multi-distillation Networkcited by 227
  8. 2018MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devicescited by 180
  9. 2017Apprentice: Using Knowledge Distillation Techniques To Improve Low-Precision Network Accuracycited by 100
  10. 2019AI Benchmark: All About Deep Learning on Smartphones in 2019cited by 76
  11. 2018Wide Activation for Efficient and Accurate Image Super-Resolutioncited by 64
  12. 2020PAMS: Quantized Super-Resolution via Parameterized Max Scalecited by 46
  13. 2020AIM 2020 Challenge on Efficient Super-Resolution: Methods and Resultscited by 36
  14. 2021Real-Time Quantized Image Super-Resolution on Mobile NPUs, Mobile AI 2021 Challenge: Reportcited by 34
  15. 2016Deep Roots: Improving CNN Efficiency with Hierarchical Filter Groupscited by 32
  16. 2018A Quantization-Friendly Separable Convolution for MobileNetscited by 22
  17. 2019AIM 2019 Challenge on Constrained Super-Resolution: Methods and Resultscited by 18

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 10; 2023: 9; 2024: 4; 2025: 3; 2026: 2.

30 papers.

Sort:

  1. 2022NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Resultscited by 40
  2. 2022CADyQ: Content-Aware Dynamic Quantization for Image Super-Resolutioncited by 26
  3. 2022Hitchhiker's Guide to Super-Resolution: Introduction and Recent Advancescited by 21
  4. 2022Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Reportcited by 16
  5. 2023Reconstructed Convolution Module Based Look-Up Tables for Efficient Image Super-Resolutioncited by 15
  6. 2021edge-SR: Super-Resolution For The Massescited by 9
  7. 2024RefQSR: Reference-based Quantization for Image Super-Resolution Networkscited by 8
  8. 2024AdaBM: On-the-Fly Adaptive Bit Mapping for Image Super-Resolutioncited by 8
  9. 2022Fast Nearest Convolution for Real-Time Efficient Image Super-Resolutioncited by 6
  10. 2023Bicubic++: Slim, Slimmer, Slimmest -- Designing an Industry-Grade Super-Resolution Networkcited by 6
  11. 2024Dynamic Contrastive Knowledge Distillation for Efficient Image Restorationcited by 6
  12. 2023SYENet: A Simple Yet Effective Network for Multiple Low-Level Vision Tasks with Real-time Performance on Mobile Devicecited by 5
18 more
  1. 2023QuickSRNet: Plain Single-Image Super-Resolution Architecture for Faster Inference on Mobile Platformscited by 4
  2. 2023Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networkscited by 4
  3. 2022IMDeception: Grouped Information Distilling Super-Resolution Networkcited by 3
  4. 2022XCAT -- Lightweight Quantized Single Image Super-Resolution using Heterogeneous Group Convolutions and Cross Concatenationcited by 3
  5. 2022Real-Time Super-Resolution for Real-World Images on Mobile Devicescited by 2
  6. 2022Efficient Multi-Purpose Cross-Attention Based Image Alignment Block for Edge Devicescited by 1
  7. 2022NAWQ-SR: A Hybrid-Precision NPU Engine for Efficient On-Device Super-Resolutioncited by 1
  8. 2023Efficient neural supersampling on a novel gaming datasetcited by 1
  9. 2023ACNPU: A 4.75TOPS/W 1080P@30FPS Super Resolution Accelerator with Decoupled Asymmetric Convolutioncited by 1
  10. 2024Task-driven single-image super-resolution reconstruction of document scanscited by 1
  11. 2026Decoder-Free Distillation for Quantized Image Restorationcited by 1
  12. 2021Machine Learning Advances aiding Recognition and Classification of Indian Monuments and Landmarkscited by 0
  13. 2023Towards Clip-Free Quantized Super-Resolution Networks: How to Tame Representative Imagescited by 0
  14. 2023TMSR: Tiny Multi-path CNNs for Super Resolutioncited by 0
  15. 2025Compressed Domain Prior-Guided Video Super-Resolution for Cloud Gaming Contentcited by 0
  16. 2025Crafting Query-Aware Selective Attention for Single Image Super-Resolutioncited by 0
  17. 2025Task-driven real-world super-resolution of document scanscited by 0
  18. 2026Exploring Real-Time Super-Resolution: Benchmarking and Fine-Tuning for Streaming Contentcited by 0