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

arXiv: https://arxiv.org/abs/1707.02937 · HTML: https://mlsanity.com/paper/1707.02937/

## Key facts

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

## Built on by (most cited first, 30 of 37)

- [Deep Burst Super-Resolution](https://mlsanity.com/paper/2101.10997/) (2021)
- [Adversarial Open Domain Adaptation for Sketch-to-Photo Synthesis](https://mlsanity.com/paper/2104.05703/) (2021)
- [Beyond Joint Demosaicking and Denoising: An Image Processing Pipeline for a Pixel-bin Image Sensor](https://mlsanity.com/paper/2104.09398/) (2021)
- [Single-Image HDR Reconstruction by Multi-Exposure Generation](https://mlsanity.com/paper/2210.15897/) (2022)
- [Data Redaction from Pre-trained GANs](https://mlsanity.com/paper/2206.14389/) (2022)
- [CamLessMonoDepth: Monocular Depth Estimation with Unknown Camera Parameters](https://mlsanity.com/paper/2110.14347/) (2021)
- [AtmoDist: Self-supervised Representation Learning for Atmospheric Dynamics](https://mlsanity.com/paper/2202.01897/) (2022)
- [Improving Feature Stability during Upsampling -- Spectral Artifacts and the Importance of Spatial Context](https://mlsanity.com/paper/2311.17524/) (2023)
- [Rethinking Cross-Generator Image Forgery Detection through DINOv3](https://mlsanity.com/paper/2511.22471/) (2025)
- [ShuffleUNet: Super resolution of diffusion-weighted MRIs using deep learning](https://mlsanity.com/paper/2102.12898/) (2021)
- [An Energy-Efficient Edge Computing Paradigm for Convolution-based Image Upsampling](https://mlsanity.com/paper/2107.07647/) (2021)
- [Addressing materials' microstructure diversity using transfer learning](https://mlsanity.com/paper/2107.13841/) (2021)
- [DeformIrisNet: An Identity-Preserving Model of Iris Texture Deformation](https://mlsanity.com/paper/2207.08980/) (2022)
- [MIPI 2023 Challenge on RGB+ToF Depth Completion: Methods and Results](https://mlsanity.com/paper/2304.13916/) (2023)
- [Sapiens2](https://mlsanity.com/paper/2604.21681/) (2026)
- [Perceptual cGAN for MRI Super-resolution](https://mlsanity.com/paper/2201.09314/) (2022)
- [FreqU-FNet: Frequency-Aware U-Net for Imbalanced Medical Image Segmentation](https://mlsanity.com/paper/2505.17544/) (2025)
- [Upsampling layers for music source separation](https://mlsanity.com/paper/2111.11773/) (2021)
- [HIME: Efficient Headshot Image Super-Resolution with Multiple Exemplars](https://mlsanity.com/paper/2203.14863/) (2022)
- [Surround-View Cameras based Holistic Visual Perception for Automated Driving](https://mlsanity.com/paper/2206.05542/) (2022)
- [DarkDeblur: Learning single-shot image deblurring in low-light condition](https://mlsanity.com/paper/2503.02194/) (2025)
- [Super-Resolved Canopy Height Mapping from Sentinel-2 Time Series Using Airborne LiDAR HD Reference Data across Metropolitan France](https://mlsanity.com/paper/2512.11524/) (2025)
- [UNeRF: Time and Memory Conscious U-Shaped Network for Training Neural Radiance Fields](https://mlsanity.com/paper/2206.11952/) (2022)
- [Discover the Mysteries of the Maya: Selected Contributions from the Machine Learning Challenge & The Discovery Challenge Workshop at ECML PKDD 2021](https://mlsanity.com/paper/2208.03163/) (2022)
- [Localization of Ultra-dense Emitters with Neural Networks](https://mlsanity.com/paper/2305.05542/) (2023)
- [Dynamic Deep Learning Based Super-Resolution For The Shallow Water Equations](https://mlsanity.com/paper/2404.06400/) (2024)
- [GANeXt: A Fully ConvNeXt-Enhanced Generative Adversarial Network for MRI- and CBCT-to-CT Synthesis](https://mlsanity.com/paper/2512.19336/) (2025)
- [Spatio-Temporal Wildfire Spread Prediction in Canada using a Video Swin-Hybrid-U-Net and Satellite Imagery](https://mlsanity.com/paper/2606.20693/) (2026)
- [Advances in Electron Microscopy with Deep Learning](https://mlsanity.com/paper/2101.01178/) (2021)
- [Towards Representation Learning for Atmospheric Dynamics](https://mlsanity.com/paper/2109.09076/) (2021)

## 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.