# Assessing Image Inpainting via Re-Inpainting Self-Consistency Evaluation

arXiv: https://arxiv.org/abs/2405.16263 · HTML: https://mlsanity.com/paper/2405.16263/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 31 references that are in the map.
- First posted on arXiv 25 May 2024.

## Builds on (most cited first, 12 of 12)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [MUSIQ: Multi-scale Image Quality Transformer](https://mlsanity.com/paper/2108.05997/) (2021)
- [Resolution-robust Large Mask Inpainting with Fourier Convolutions](https://mlsanity.com/paper/2109.07161/) (2021)
- [Free-Form Image Inpainting with Gated Convolution](https://mlsanity.com/paper/1806.03589/) (2018)
- [Large Scale Image Completion via Co-Modulated Generative Adversarial Networks](https://mlsanity.com/paper/2103.10428/) (2021)
- [EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning](https://mlsanity.com/paper/1901.00212/) (2019)
- [SC-FEGAN: Face Editing Generative Adversarial Network with User's Sketch and Color](https://mlsanity.com/paper/1902.06838/) (2019)
- [Bringing Old Photos Back to Life](https://mlsanity.com/paper/2004.09484/) (2020)
- [Inst-Inpaint: Instructing to Remove Objects with Diffusion Models](https://mlsanity.com/paper/2304.03246/) (2023)
- [Relation-Aware Diffusion Model for Controllable Poster Layout Generation](https://mlsanity.com/paper/2306.09086/) (2023)

## Built on by (most cited first, 1 of 1)

- [WINGS: Reference-Free Gaussian Splatting Inpainting with 3D-Native Generative Priors](https://mlsanity.com/paper/2609.37816/) (2026)

## Abstract (arXiv)

Image inpainting, the task of reconstructing missing segments in corrupted images using available data, faces challenges in ensuring consistency and fidelity, especially under information-scarce conditions. Traditional evaluation methods, heavily dependent on the existence of unmasked reference images, inherently favor certain inpainting outcomes, introducing biases. Addressing this issue, we introduce an innovative evaluation paradigm that utilizes a self-supervised metric based on multiple re-inpainting passes. This approach, diverging from conventional reliance on direct comparisons in pixel or feature space with original images, emphasizes the principle of self-consistency to enable the exploration of various viable inpainting solutions, effectively reducing biases. Our extensive experiments across numerous benchmarks validate the alignment of our evaluation method with human judgment.