mlsanityData to 14 Oct 2026

Checkerboard artifact free sub-pixel convolution: A note on sub-pixel convolution, resize convolution and convolution resize

arXiv 1707.02937 ↗First posted 10 Jul 2017cs.CVcited by 37

Not on the map. Why?

Abstract (arXiv)

The most prominent problem associated with the deconvolution layer is the presence of checkerboard artifacts in output images and dense labels. To combat this problem, smoothness constraints, post processing and different architecture designs have been proposed. Odena et al. highlight three sources of checkerboard artifacts: deconvolution overlap, random initialization and loss functions. In this note, we proposed an initialization method for sub-pixel convolution known as convolution NN resize. Compared to sub-pixel convolution initialized with schemes designed for standard convolution kernels, it is free from checkerboard artifacts immediately after initialization. Compared to resize convolution, at the same computational complexity, it has more modelling power and converges to solutions with smaller test errors.

Key facts. Cited by 37 papers in the map (data through 14 Oct 2026). First posted on arXiv 10 Jul 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 10; 2022: 10; 2023: 6; 2024: 3; 2025: 5; 2026: 3.

37 papers.

Sort:

  1. 2021Deep Burst Super-Resolutioncited by 83
  2. 2021Adversarial Open Domain Adaptation for Sketch-to-Photo Synthesiscited by 24
  3. 2021Beyond Joint Demosaicking and Denoising: An Image Processing Pipeline for a Pixel-bin Image Sensorcited by 22
  4. 2022Single-Image HDR Reconstruction by Multi-Exposure Generationcited by 20
  5. 2022Data Redaction from Pre-trained GANscited by 18
  6. 2021CamLessMonoDepth: Monocular Depth Estimation with Unknown Camera Parameterscited by 8
  7. 2022AtmoDist: Self-supervised Representation Learning for Atmospheric Dynamicscited by 8
  8. 2023Improving Feature Stability during Upsampling -- Spectral Artifacts and the Importance of Spatial Contextcited by 8
  9. 2025Rethinking Cross-Generator Image Forgery Detection through DINOv3cited by 8
  10. 2021ShuffleUNet: Super resolution of diffusion-weighted MRIs using deep learningcited by 4
  11. 2021An Energy-Efficient Edge Computing Paradigm for Convolution-based Image Upsamplingcited by 4
  12. 2021Addressing materials' microstructure diversity using transfer learningcited by 4
25 more
  1. 2022DeformIrisNet: An Identity-Preserving Model of Iris Texture Deformationcited by 4
  2. 2023MIPI 2023 Challenge on RGB+ToF Depth Completion: Methods and Resultscited by 4
  3. 2026Sapiens2cited by 4
  4. 2022Perceptual cGAN for MRI Super-resolutioncited by 3
  5. 2025FreqU-FNet: Frequency-Aware U-Net for Imbalanced Medical Image Segmentationcited by 3
  6. 2021Upsampling layers for music source separationcited by 2
  7. 2022HIME: Efficient Headshot Image Super-Resolution with Multiple Exemplarscited by 2
  8. 2022Surround-View Cameras based Holistic Visual Perception for Automated Drivingcited by 2
  9. 2025DarkDeblur: Learning single-shot image deblurring in low-light conditioncited by 2
  10. 2025Super-Resolved Canopy Height Mapping from Sentinel-2 Time Series Using Airborne LiDAR HD Reference Data across Metropolitan Francecited by 2
  11. 2022UNeRF: Time and Memory Conscious U-Shaped Network for Training Neural Radiance Fieldscited by 1
  12. 2022Discover the Mysteries of the Maya: Selected Contributions from the Machine Learning Challenge & The Discovery Challenge Workshop at ECML PKDD 2021cited by 1
  13. 2023Localization of Ultra-dense Emitters with Neural Networkscited by 1
  14. 2024Dynamic Deep Learning Based Super-Resolution For The Shallow Water Equationscited by 1
  15. 2025GANeXt: A Fully ConvNeXt-Enhanced Generative Adversarial Network for MRI- and CBCT-to-CT Synthesiscited by 1
  16. 2026Spatio-Temporal Wildfire Spread Prediction in Canada using a Video Swin-Hybrid-U-Net and Satellite Imagerycited by 1
  17. 2021Advances in Electron Microscopy with Deep Learningcited by 0
  18. 2021Towards Representation Learning for Atmospheric Dynamicscited by 0
  19. 2022Rethinking Generative Methods for Image Restoration in Physics-based Vision: A Theoretical Analysis from the Perspective of Informationcited by 0
  20. 2023Novel Building Detection and Location Intelligence Collection in Aerial Satellite Imagerycited by 0
  21. 2023ParGANDA: Making Synthetic Pedestrians A Reality For Object Detectioncited by 0
  22. 2023Improved Data Generation for Enhanced Asset Allocation: A Synthetic Dataset Approach for the Fixed Income Universecited by 0
  23. 2024Deep-Motion-Net: GNN-based volumetric organ shape reconstruction from single-view 2D projectionscited by 0
  24. 2024Distribution Learning with Valid Outputs Beyond the Worst-Casecited by 0
  25. 2026Layer Selection in Feature-Based Losses Affects Image Quality and Microstructural Consistency in Deep Learning Super-Resolution of Brain Diffusion MRIcited by 0