# Toward Bridging the Simulated-to-Real Gap: Benchmarking Super-Resolution on Real Data

arXiv: https://arxiv.org/abs/1809.06420 · HTML: https://mlsanity.com/paper/1809.06420/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Sep 2018.

## Built on by (most cited first, 7 of 7)

- [ERQA: Edge-Restoration Quality Assessment for Video Super-Resolution](https://mlsanity.com/paper/2110.09992/) (2021)
- [Multitemporal and multispectral data fusion for super-resolution of Sentinel-2 images](https://mlsanity.com/paper/2301.11154/) (2023)
- [Image Super-Resolution Using T-Tetromino Pixels](https://mlsanity.com/paper/2111.09013/) (2021)
- [Real-World Single Image Super-Resolution Under Rainy Condition](https://mlsanity.com/paper/2206.08345/) (2022)
- [Towards Super-Resolution CEST MRI for Visualization of Small Structures](https://mlsanity.com/paper/2112.01905/) (2021)
- [The Best of Both Worlds: a Framework for Combining Degradation Prediction with High Performance Super-Resolution Networks](https://mlsanity.com/paper/2211.05018/) (2022)
- [Real-World Video for Zoom Enhancement based on Spatio-Temporal Coupling](https://mlsanity.com/paper/2306.13875/) (2023)

## Abstract (arXiv)

Capturing ground truth data to benchmark super-resolution (SR) is challenging. Therefore, current quantitative studies are mainly evaluated on simulated data artificially sampled from ground truth images. We argue that such evaluations overestimate the actual performance of SR methods compared to their behavior on real images. Toward bridging this simulated-to-real gap, we introduce the Super-Resolution Erlangen (SupER) database, the first comprehensive laboratory SR database of all-real acquisitions with pixel-wise ground truth. It consists of more than 80k images of 14 scenes combining different facets: CMOS sensor noise, real sampling at four resolution levels, nine scene motion types, two photometric conditions, and lossy video coding at five levels. As such, the database exceeds existing benchmarks by an order of magnitude in quality and quantity. This paper also benchmarks 19 popular single-image and multi-frame algorithms on our data. The benchmark comprises a quantitative study by exploiting ground truth data and qualitative evaluations in a large-scale observer study. We also rigorously investigate agreements between both evaluations from a statistical perspective. One interesting result is that top-performing methods on simulated data may be surpassed by others on real data. Our insights can spur further algorithm development, and the publicy available dataset can foster future evaluations.