---
license: other
task_categories:
  - image-to-text
  - text-to-image
tags:
  - art
  - chinese-painting
  - museum
  - caption
  - ocr
  - parquet
size_categories:
  - 10K<n<100K
---

# Chinese Painting Collection

91,438 images of Chinese paintings with bilingual (Chinese/English) VLM captions,
calligraphy OCR transcriptions, view-type classification, a long caption on part
of the set, and foreign-object detection with one final crop box per image.

## Sources

| Component | Source | Images | Image license |
|---|---|---|---|
| `npm_tw_c0`–`npm_tw_c3` | [National Palace Museum (Taipei) Open Data](https://theme.npm.edu.tw/opendata/) | 86,666 | Taiwan Open Government Data License v1 (attribution required) |
| `met_china` | [The Metropolitan Museum of Art Open Access](https://www.metmuseum.org/art/collection) | 1,186 | CC0 |
| `fsg` | [Smithsonian Freer Gallery of Art / National Museum of Asian Art](https://asia.si.edu/) | 1,782 | CC0 |
| `princeton` | [Princeton University Art Museum](https://artmuseum.princeton.edu/) | 1,631 | Check source terms; intended for research use |
| `aic_china` | [Art Institute of Chicago](https://www.artic.edu/collection) | 173 | CC0 |

The four NPM components are quarterly shards of the NPM Open Data release
(`c0`–`c3`), deduplicated and filtered to usable views. The museum components
were harvested through each institution's public API and filtered to
China-related Chinese-painting objects.

## Production pipeline

1. **Harvest** — images and tombstone metadata (title, artist, date, object
   number) fetched from the source APIs / open-data packages.
2. **Rule-based cleaning** — removal of non-artwork views: rolled scrolls,
   installation shots, covers, damaged or unreadable images, and near-duplicates.
   Only `full`, `mounted`, and `detail` views are kept.
3. **VLM labeling** (gemini-3.5-flash-lite) — for every kept image:
   - `caption_zh` / `caption_en`: free-form descriptions of the scene,
     composition, style, and medium. Captions were instructed to avoid boilerplate
     openings and to be specific.
   - `ocr_text`: transcription of visible Chinese calligraphy (inscriptions,
     colophons) and a description of seal impressions where present.
   - `view_type`: `full` (whole artwork), `mounted` (artwork with mounting
     visible), or `detail` (close-up crop).
   - `culture`: `chinese` (88,928), `japanese` (2,255), `korean` (83),
     `western` (10), `other` (20), `uncertain` (142). Non-Chinese items are a
     small minority kept for completeness; filter on this field if unwanted.
4. **Foreign-object detection** — museum photography of NPM flatworks often
   includes color charts, rulers, accession labels, or the surrounding desk.
   A detection pass lists what it sees (`artifacts`: `label`, `color_chart`,
   `ruler`, `desk`, `glare`, `other`; non-empty for 39,509 images) and returns a
   crop box for the artwork.
5. **Second cropping round** — for the images the first pass flagged, the model
   was shown the already-cropped view and asked for the artwork box *inside* it.
   The two boxes are composed into one box in the coordinates of the original
   photograph, which is what `bbox` holds. There is exactly one `bbox` per row
   and it is the final one: present for 39,338 images, empty for the rest, which
   means the whole frame is the artwork. 171 flagged images have no box; they are
   kept uncropped rather than discarded.
6. **Text folded into the captions** — the transcription from step 3 was moved
   into `caption_zh` / `caption_en` as a trailing block, because the caption is
   what a text-to-image model consumes. Catalogue-label lines (accession numbers,
   album titles) were dropped in the process, since the crop removes that strip
   from the training view. `ocr_text` is kept as the raw field.
7. **Long captions** — 59,860 of the 91,438 images received a second, much longer
   caption (`caption_long`) written by a different model (DeepSeek V4.1 Flash,
   reasoning disabled), describing the same image in more detail. Lengths below
   are counted with the training tokenizer: median 488 tokens, p90 817, max
   4,415. 616 candidates were discarded for repeating a sentence; length alone
   was never a reason to discard. The model was shown the cropped view (the
   `bbox` crop), i.e. what a trainer sees, not the raw photograph. Existing
   captions were not replaced, so a row can carry up to three captions.

## Data format

Parquet shards under `default/train/part-*.parquet`, one row per image. The
photograph is inside the row in an `image` column of type
`struct<bytes: binary, path: string>`, the layout the `datasets` library uses for
its `Image` feature, so every image can be previewed in the dataset viewer:

```python
from datasets import load_dataset
ds = load_dataset("kaupane/chinese-painting-collection", split="train")
ds[0]["image"].size          # the photograph
ds[0]["caption_long"]        # long caption, or None
```

Columns: `image_id`, `shard`, `source`, `object_no`, `title`, `artist`,
`category`, `culture`, `view_type`, `caption_zh`, `caption_en`, `caption_long`,
`ocr_text`, `artifacts` (list), `bbox` (list of four floats), `image`.

`bbox` is a normalised `[x_min, y_min, x_max, y_max]` in per-mille coordinates
(0–1000). Crop with
`img.crop((x0*W//1000, y0*H//1000, x1*W//1000, y1*H//1000))`; when it is empty,
use the image as it is. `previews/` shows twelve before/after pairs so the effect
of the crop can be seen without loading the data.

View-type distribution: `mounted` 43,548 · `full` 36,253 · `detail` 11,637.

## Suggested filtering

- **Training on artwork only**: apply the `bbox` crop (recommended). Images with
  an empty `bbox` need no crop.
- **Cleanest subset**: also drop rows whose `artifacts` is non-empty.
- **Exclude mounting borders**: keep `view_type == "full"` or `"detail"`.
- **Chinese paintings only**: keep `culture == "chinese"`.
- **Long-text training**: use `caption_long` where present.

## Caveats

- Captions and OCR are machine-generated by lightweight VLMs; expect occasional
  errors, especially in cursive-script calligraphy transcription. Where a
  transcription is quoted in a caption, it is the model's reading, not a
  scholarly edition.
- The crop box is the model's answer to "where is the artwork". It usually
  removes the color chart, ruler and label strip, and in most cases leaves a
  thin strip of the photograph's background along one edge.
- `detail` views are close-ups of larger works and may duplicate content present
  in `full` views of the same object (same `object_no`).
- A small number of images have artifacts but no box (171) and are shipped
  uncropped.

## Licensing

Images retain the licenses of their sources (see table above). The NPM portion
requires attribution to 國立故宮博物院 (National Palace Museum, Taipei) under
the Taiwan Open Government Data License v1. Annotations (captions, OCR,
classifications, boxes) are released under CC-BY-4.0.