mlsanityData to 7 Oct 2026

Thera: Aliasing-Free Arbitrary-Scale Super-Resolution with Neural Heat Fields

arXiv 2311.17643 ↗First posted 29 Nov 2023cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

Recent approaches to arbitrary-scale single image super-resolution (ASR) use neural fields to represent continuous signals that can be sampled at arbitrary resolutions. However, point-wise queries of neural fields do not naturally match the point spread function (PSF) of pixels, which may cause aliasing in the super-resolved image. Existing methods attempt to mitigate this by approximating an integral version of the field at each scaling factor, compromising both fidelity and generalization. In this work, we introduce neural heat fields, a novel neural field formulation that inherently models a physically exact PSF. Our formulation enables analytically correct anti-aliasing at any desired output resolution, and -- unlike supersampling -- at no additional cost. Building on this foundation, we propose Thera, an end-to-end ASR method that substantially outperforms existing approaches, while being more parameter-efficient and offering strong theoretical guarantees. The project page is at https://therasr.github.io.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 47 of its 62 references that are in the map. First posted on arXiv 29 Nov 2023.

Builds on

47 of 62 references.

References · 47

  1. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  2. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  3. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,698
  4. 2022A ConvNet for the 2020scited by 3,545
  5. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  6. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  7. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  8. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  9. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,013
  10. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,957
  11. 2021SwinIR: Image Restoration Using Swin Transformercited by 1,869
  12. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
35 more
  1. 2021Image Super-Resolution via Iterative Refinementcited by 1,506
  2. 2015Spatial Transformer Networkscited by 1,423
  3. 2021Alias-Free Generative Adversarial Networkscited by 1,181
  4. 2016HyperNetworkscited by 1,003
  5. 2018Residual Dense Network for Image Super-Resolutioncited by 796
  6. 2020Convolutional Occupancy Networkscited by 760
  7. 2023Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fieldscited by 667
  8. 2017The Perception-Distortion Tradeoffcited by 582
  9. 2020Learning Continuous Image Representation with Local Implicit Image Functioncited by 550
  10. 2021Neural Fields in Visual Computing and Beyondcited by 539
  11. 2018fastMRI: An Open Dataset and Benchmarks for Accelerated MRIcited by 464
  12. 2015Sketch-based Manga Retrieval using Manga109 Datasetcited by 326
  13. 2021ICON: Implicit Clothed humans Obtained from Normalscited by 283
  14. 2023Implicit Diffusion Models for Continuous Super-Resolutioncited by 216
  15. 2020MetaSDF: Meta-learning Signed Distance Functionscited by 214
  16. 2013Gradient Magnitude Similarity Deviation: A Highly Efficient Perceptual Image Quality Indexcited by 193
  17. 2019Meta-SR: A Magnification-Arbitrary Network for Super-Resolutioncited by 180
  18. 2023Inversion by Direct Iteration: An Alternative to Denoising Diffusion for Image Restorationcited by 164
  19. 2021BACON: Band-limited Coordinate Networks for Multiscale Scene Representationcited by 142
  20. 2023Tri-MipRF: Tri-Mip Representation for Efficient Anti-Aliasing Neural Radiance Fieldscited by 142
  21. 2021Local Texture Estimator for Implicit Representation Functioncited by 126
  22. 2020i3DMM: Deep Implicit 3D Morphable Model of Human Headscited by 109
  23. 2022Learning Neural Parametric Head Modelscited by 73
  24. 2022CiaoSR: Continuous Implicit Attention-in-Attention Network for Arbitrary-Scale Image Super-Resolutioncited by 69
  25. 2023Neural LiDAR Fields for Novel View Synthesiscited by 65
  26. 2022ImFace: A Nonlinear 3D Morphable Face Model with Implicit Neural Representationscited by 64
  27. 2021Neural Fields as Learnable Kernels for 3D Reconstructioncited by 58
  28. 2021UltraSR: Spatial Encoding is a Missing Key for Implicit Image Function-based Arbitrary-Scale Super-Resolutioncited by 56
  29. 2015Seven ways to improve example-based single image super resolutioncited by 51
  30. 2023Cascaded Local Implicit Transformer for Arbitrary-Scale Super-Resolutioncited by 51
  31. 2023Super-Resolution Neural Operatorcited by 48
  32. 2022Neural Fourier Filter Bankcited by 29
  33. 2022JIFF: Jointly-aligned Implicit Face Function for High Quality Single View Clothed Human Reconstructioncited by 26
  34. 2023Polynomial Neural Fields for Subband Decomposition and Manipulationcited by 22
  35. 2022CUF: Continuous Upsampling Filterscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 3.

5 papers.

  1. 2025Continuous Space-Time Video Super-Resolution with 3D Fourier Fieldscited by 3
  2. 2025SuperF: Neural Implicit Fields for Multi-Image Super-Resolutioncited by 1
  3. 2026SSPFormer: Self-Supervised Pretrained Transformer for MRI Imagescited by 0
  4. 2026Frequency Decoupled Framework for Screen Content Image Super-Resolutioncited by 0
  5. 2026GB-LSR: Local Spectral Decoding with a Learned Global Bandwidth for Arbitrary-Scale Super-Resolutioncited by 0