---
license: bsd-3-clause
pretty_name: DSR Benchmark — Public Release
task_categories:
  - image-to-image
tags:
  - super-resolution
  - image-quality-assessment
  - no-reference-image-quality-assessment
  - mean-opinion-score
---

# DSR Benchmark — Public Release

DSR Benchmark is a dataset for studying the quality of images produced by
classic and diffusion-based Super-Resolution (SR) methods. The public release
contains all selected SR images, along with the LR inputs, HR source links, and
subjective quality scores for 60% of the SR images in each category. The data
associated with the remaining 40% are reserved for a future challenge.

## Dataset summary

| Component | Count |
| --- | ---: |
| HR images in the source collection | 10,036 |
| LR images: 2 scales × 4 variants | 80,288 |
| All generated SR candidates: 10,036 × 8 × 14 | 1,124,032 |
| Selected and subjectively evaluated Diffusion SR images | 10,000 |
| Selected and subjectively evaluated Classic SR images | 4,000 |
| Total selected SR images (all images are published) | 14,000 |
| SR images with public MOS values and references | 60% of each SR category |
| SR images reserved for the challenge | 40% of each SR category |

The split is performed at the source HR-image level. All SR images sharing the
same five-digit identifier `xxxxx` belong to the same partition; variants of a
single HR image are never divided between the public and hidden partitions.
The 60% target is calculated independently for Diffusion SR and Classic SR.

## HR image sources

The 10,036 HR images were selected from the following datasets:

- [KonIQ-10k](https://arxiv.org/abs/1910.06180);
- [Microsoft COCO](https://cocodataset.org/) — images from the `train`,
  `validation`, and `test` splits were used;
- [AVA: A Large-Scale Database for Aesthetic Visual Analysis](https://ieeexplore.ieee.org/document/6247954);
- FLIVE, introduced in
  [From Patches to Pictures (PaQ-2-PiQ)](https://openaccess.thecvf.com/content_CVPR_2020/html/Ying_From_Patches_to_Pictures_PaQ-2-PiQ_Mapping_the_Perceptual_Space_of_CVPR_2020_paper.html).

Each HR image was assigned a five-digit identifier. HR filenames follow this
pattern:

```text
xxxxx_vg_dsr.jpg
```

Example:

```text
00014_vg_dsr.jpg
```

The HR image files are not redistributed in the public repository. For the
public partition, links to the original HR images or their source pages are
provided in `metadata/hr_source_links.csv`.

## LR generation

Each HR image was downscaled using bicubic interpolation at two scale factors:

- `x2`;
- `x4`.

Four degradation variants were prepared for each scale factor:

| Label | LR variant |
| --- | --- |
| `jpeg` | JPEG compression |
| `blur` | blurring |
| `noise` | Gaussian noise |
| `resized` | bicubic downscaling without an additional distortion |

Therefore, each HR image has eight corresponding LR images:

$$
2\ \text{scales} \times 4\ \text{variants} = 8\ \text{LR images}.
$$

LR filenames follow this pattern:

```text
xxxxx_vg_dsr_{distortion}_{x2/x4}.jpg
```

Examples:

```text
00014_vg_dsr_blur_x2.jpg
00014_vg_dsr_noise_x4.jpg
00014_vg_dsr_resized_x4.jpg
```

## Super-Resolution methods

Ten diffusion-based and four classic SR methods were applied to each LR image.

### Diffusion SR

| Method | Dataset label |
| --- | --- |
| ResShift | `ResShift` |
| SinSR | `SinSR` |
| SeeSR | `SeeSR` |
| PASD | `PASD` |
| PASD-SDXL | `SPASD` |
| DiT4SR | `DiT4SR` |
| SUPIR | `SUPIR` |
| CODSR | `CODSR` |
| InvSR | `InvSR` |
| PassionSR | `PassionSR` |

Note that **PASD-SDXL** results are labeled `SPASD` in the filenames.

### Classic SR

| Method | Dataset label |
| --- | --- |
| HAT | `HAT` |
| Real-ESRGAN | `Real-ESRGAN` |
| SwinIR | `SwinIR` |
| RealSR | `RealSR` |

SR filenames follow this pattern:

```text
{method}_{x2/x4}_xxxxx_vg_dsr_{distortion}_{x2/x4}.{extension}
```

Examples:

```text
CODSR_x2_00014_vg_dsr_blur_x2.jpg
SPASD_x4_00014_vg_dsr_noise_x4.png
SwinIR_x4_02310_vg_dsr_resized_x4.png
```

The filename components have the following meanings:

| Component | Meaning |
| --- | --- |
| `method` | the SR method used |
| first `x2` or `x4` | the SR upscaling factor |
| `xxxxx` | identifier shared by the corresponding HR and LR content |
| `distortion` | the LR degradation variant |
| final `x2` or `x4` | the downscaling factor of the corresponding LR image |
| `extension` | the actual file format, such as `jpg` or `png` |

## Subjective evaluation

The following images were selected from the 1,124,032 generated candidates:

- 10,000 Diffusion SR images;
- 4,000 Classic SR images.

The images were evaluated using a **no-reference MOS** methodology: observers
rated the visual quality of each SR image without seeing its corresponding LR
or HR image. The subjective experiment used a scale from 1 to 5, where a
higher score indicates higher perceived quality. The MOS for an SR image is
obtained by aggregating its individual ratings.

Public annotations cover 60% of the Diffusion SR images and 60% of the Classic
SR images. For the remaining 40%, the SR images themselves are available, but
their corresponding MOS values, individual votes, LR images, and HR source
links are not published.

> **Note on score representation.** The CSV files contain numeric values in
> the form in which they were exported and used to calculate MOS. The data
> collection procedure used a 1–5 scale. When using these files, take into
> account the score encoding employed by the particular dataset version.

## Public repository structure

To stay within repository limits, image files are sharded by the first three
digits of their five-digit source identifier. For example, all files associated
with source ID `00014` are stored in shard `000`. The shard name is denoted by
`{prefix}` below and is always equal to `xxxxx[:3]`.

```text
.
├── README.md
├── images/
│   ├── lr/
│   │   ├── x2/
│   │   │   └── {prefix}/
│   │   │       └── xxxxx_vg_dsr_{distortion}_x2.jpg
│   │   └── x4/
│   │       └── {prefix}/
│   │           └── xxxxx_vg_dsr_{distortion}_x4.jpg
│   └── sr/
│       ├── diffusion/
│       │   └── {prefix}/
│       │       └── {method}_{scale}_xxxxx_vg_dsr_{distortion}_{scale}.{extension}
│       └── classic/
│           └── {prefix}/
│               └── {method}_{scale}_xxxxx_vg_dsr_{distortion}_{scale}.{extension}
└── metadata/
    ├── mos.csv
    ├── votes.csv
    ├── mos_csr.csv
    ├── votes_csr.csv
    └── hr_source_links.csv
```

The public repository intentionally does not include `split_assignment.csv` or
`sr_manifest.csv`. Available SR and LR images, as well as the corresponding HR
source links, can be matched using the five-digit `xxxxx` identifier.

### `metadata/hr_source_links.csv`

Contains source information and links for the HR references associated with
the public partition. The HR files themselves are not redistributed in this
repository. Match a row to an LR or SR image using `source_id` or
`hr_file_name`, then use `source_url` to access the original image or its
source page.

The `source_url_type` column indicates the type of destination:

- `exact_image` — a direct link to the original image;
- `exact_image_page` — a page containing the original image;
- `dataset_page_only` — the source dataset page, when an image-specific link
  is unavailable;
- `unresolved` — no suitable public link was resolved.

### `images/lr/x2/{prefix}/` and `images/lr/x4/{prefix}/`

Contain the LR inputs for the public partition, organized by downscaling
factor. LR images corresponding to the hidden 40% of SR images are not
included.

### `images/sr/diffusion/{prefix}/`

Contains all 10,000 selected Diffusion SR images, including images from both
the public and challenge partitions.

### `images/sr/classic/{prefix}/`

Contains all 4,000 selected Classic SR images, including images from both the
public and challenge partitions.

## Annotation files

### `metadata/mos_diffusion_SR.csv`

Contains MOS values for public Diffusion SR images.

```csv
,img_name,mos
0,CODSR_x2_00014_vg_dsr_blur_x2.jpg,1.0
```

| Column | Description |
| --- | --- |
| first unnamed column | row index added during export; it can be ignored |
| `img_name` | filename of the Diffusion SR image |
| `mos` | aggregated subjective quality score for the image |

### `metadata/mos_classic_SR.csv`

Uses the same structure as `mos_diffusion_SR.csv`, but contains MOS values for public
Classic SR images.

### `metadata/votes_diffusion_SR.csv`

Contains individual ratings for public Diffusion SR images.

```csv
,INPUT:image_a,OUTPUT:result,ASSIGNMENT:worker_id
0,CODSR_x2_00014_vg_dsr_blur_x2.jpg,1.0,db490266783ff3f7b7cbfe29cf9cbffa
```

| Column | Description |
| --- | --- |
| first unnamed column | row index added during export; it can be ignored |
| `INPUT:image_a` | filename of the evaluated SR image |
| `OUTPUT:result` | individual numeric rating |
| `ASSIGNMENT:worker_id` | technical participant identifier stored by the annotation system |

### `metadata/votes_classic_SR.csv`

Uses the same structure as `votes_diffusion_SR.csv`, but contains individual ratings for
public Classic SR images.

## Matching SR, LR, and HR images

Files correspond to one another through the `xxxxx` identifier. For example,
the SR image

```text
CODSR_x2_00014_vg_dsr_blur_x2.jpg
```

has the following public LR reference:

```text
images/lr/x2/000/00014_vg_dsr_blur_x2.jpg
```

The corresponding HR filename is `00014_vg_dsr.jpg`. Its source link can be
obtained from the row with `source_id=00014` in
`metadata/hr_source_links.csv`.

If the corresponding LR file and HR-link row are absent from the public
repository, the SR image belongs to the challenge partition, and its MOS and
votes are not published either.

## Loading the annotations

```python
from pathlib import Path

import pandas as pd

root = Path("/path/to/dataset")

diffusion_mos = pd.read_csv(
    root / "metadata/mos.csv",
    index_col=0,
)
classic_mos = pd.read_csv(
    root / "metadata/mos_csr.csv",
    index_col=0,
)
hr_links = pd.read_csv(
    root / "metadata/hr_source_links.csv",
    dtype={"source_id": str},
)

print(diffusion_mos.head())
print(classic_mos.head())
print(hr_links.head())
```

The LR and SR paths and the HR source URL can be obtained from the source
identifier:

```python
source_id = "00014"
prefix = source_id[:3]

hr_row = hr_links.loc[hr_links["source_id"] == source_id].iloc[0]
hr_url = hr_row["source_url"]
lr_path = root / "images/lr/x2" / prefix / f"{source_id}_vg_dsr_blur_x2.jpg"
sr_path = (
    root
    / "images/sr/diffusion"
    / prefix
    / f"CODSR_x2_{source_id}_vg_dsr_blur_x2.jpg"
)
```

The source-content identifier can be extracted from a filename with a regular
expression:

```python
import re

SOURCE_ID_RE = re.compile(r"(?<!\d)(\d{5})_vg_dsr")

name = "CODSR_x2_00014_vg_dsr_blur_x2.jpg"
source_id = SOURCE_ID_RE.search(name).group(1)
print(source_id)  # 00014
```

## Recommended use cases

- training and evaluating no-reference IQA models for SR;
- comparing the perceived quality of classic and diffusion-based SR methods;
- analyzing the effects of degradation type and scale factor;
- studying the generalization of quality metrics across SR method families;
- preparing for the challenge with hidden references and subjective scores.

## Limitations

- MOS reflects the subjective perception of the participants and is not an
  objective measure of pixel-level fidelity or semantic accuracy.
- Diffusion-based methods may generate plausible details that were not present
  in the LR or HR images.
- The content distribution inherits the characteristics and potential biases
  of KonIQ-10k, COCO, AVA, and FLIVE.
- The public and challenge partitions are grouped by HR content; therefore,
  exactly 60% of the HR groups need not be public.
- Missing LR, HR source-link, or MOS data in the public release are part of the
  challenge protocol and do not indicate dataset corruption.

## Terms of use and attribution

The source images originate from KonIQ-10k, COCO, AVA, and FLIVE. The terms
and restrictions of the respective source datasets continue to apply. This
repository does not grant additional rights to the source images and does not
replace the licensing and attribution requirements established by their
authors.

When using this dataset, users should also consider the licenses and terms of
the SR method implementations used to produce the derived images.

## Citation information

```
@misc{borisov2026srgroundimagequalitygrounding,
      title={SR-Ground: Image Quality Grounding for Super-Resolved Content}, 
      author={Artem Borisov and Evgeney Bogatyrev and Khaled Abud and Dmitriy Vatolin},
      year={2026},
      eprint={2605.21244},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2605.21244}, 
}
```