Datasets:
data/ffhq_recaptioned.parquet (70,000 faces: caption clean + raw, tags, lighting) + the card
351f8c2 verified | license: cc-by-nc-sa-4.0 | |
| task_categories: | |
| - image-to-text | |
| language: | |
| - en | |
| pretty_name: FFHQ, recaptioned | |
| size_categories: | |
| - 10K<n<100K | |
| tags: | |
| - faces | |
| - portraits | |
| - captions | |
| - booru-tags | |
| - lighting | |
| - ffhq | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/ffhq_recaptioned.parquet | |
| # FFHQ, recaptioned | |
| New captions, booru tags and lighting readings for the 70,000 faces of [FFHQ](https://github.com/NVlabs/ffhq-dataset) | |
| (Flickr-Faces-HQ, NVIDIA: Karras, Laine and Aila, 2019; aligned and cropped photos of 1024x1024 from Flickr). Each row is keyed by | |
| FFHQ's own image number; this repository holds none of its images or metadata. Get the images from | |
| [FFHQ's page](https://github.com/NVlabs/ffhq-dataset), under its terms. | |
| `data/ffhq_recaptioned.parquet` (the default config): one row a face, 70,000 rows. | |
| | field | meaning | | |
| |---|---| | |
| | `image`, `file` | FFHQ's image number (00000-69999) and the file name it was read under | | |
| | `caption` | the caption with its statements of absent items removed (below) | | |
| | `caption_raw` | the caption as the model wrote it | | |
| | `new_tokens`, `truncated` | its length in tokens; whether it reached the 160-token cap | | |
| | `tags_kept` | the kept booru tags as one line | | |
| | `tags` | `[name, score, category, precision]` for every tag, as a JSON string | | |
| | `shot_light` | the lighting values the readings support, comma-separated (empty when none does) | | |
| | `luminance` | the photo's mean luminance (Rec. 709, 0-1) | | |
| | `light` | CinemaCLIP's lighting heads (cast, contrast, direction, edge, silhouette): every label's score and the labels over the model's own thresholds, as a JSON string | | |
| `raw/` keeps the runs as they were written: | |
| | folder | rows | what | | |
| |---|---|---| | |
| | `raw/20261105_085021_ffhq_captions` | 70,000 | every face: the caption and the tags | | |
| | `raw/20261105_130209_ffhq_lighting` | 70,000 | every face: CinemaCLIP's lighting heads, the luminance and the shot light values under the first rule (below) | | |
| | `raw/20261105_083559_ffhq_captions_smoke100` | 100 | the first test, faces 00000-00099 | | |
| | `raw/20261105_083756_ffhq_lighting_smoke100` | 100 | the same faces' lighting | | |
| | `raw/20261105_083805_ffhq_shotvl_light_smoke100` | 100 | a second lighting reader on the same faces: ShotVL-7B's open answer to its benchmark's question "What's the lighting of this shot?" (not used: it said "soft, from the side" of frontal flash) | | |
| | `raw/20261105_084828_ffhq_captions_askB_smoke100`, `..._askC_...`, `..._askD_...` | 100 each | three reworded asks on the same faces (below) | | |
| **Captions.** [Qwen3.5-4B](https://huggingface.co/Qwen/Qwen3.5-4B), greedy, thinking off, 160 new tokens at most, asked: | |
| > Describe this picture in one or two sentences: who is in it (apparent age, expression, hair, and any eyewear, headwear or | |
| > jewelry), each visible garment with its color and any print or text on it, the background and the light. Answer with the visible | |
| > description ONLY. Do not describe print or text if none is present. | |
| Asked for "any eyewear, headwear or jewelry" and "any print or text", the model also names what is absent ("no eyewear, headwear, | |
| or jewelry", "with no visible print or text"): 68 and 36 of the first 100 captions. Three reworded asks on the same faces did not | |
| remove it (the best still wrote one in 55 of 100, with longer captions and fewer accessories named), so `caption` drops those | |
| statements with `absences.drop` from qwen-symbolic-thinking 0.1.52 and `caption_raw` keeps the original: 58,032 captions changed, | |
| 1,178 still name an absent item, most of them exceptions that name what is there ("no jewelry beyond the earrings"), kept on | |
| purpose. Other negatives stay ("no visible smile"). Median length 65 words; 290 captions reached the 160-token cap | |
| (`truncated`). | |
| **Tags.** `animetimm/eva_giant_patch14_560.dbv4-full`, every tag with its score, category and published precision; `tags_kept` | |
| applies a leave-out list (`short hair`, `see-through`, `see-through sleeves`, `see-through shirt`, `see-through clothes`, `hair bun`, `single hair bun`, `double bun`, `parted lips`, `lips`, `realistic`, `photorealistic`, `asian`, `old`, `cosplay`) and a precision floor of 0.4. The tags are a drawing vocabulary read from photographs: | |
| `1girl` / `1boy` count people, and some tags fire on ordinary photos: `cosplay` on everyday photos (19 of the first 100 faces) and `old` on young adults (10 of 100), so both are left out of `tags_kept` and stay in `tags`. | |
| **Lighting.** [CinemaCLIP](https://huggingface.co/OZU-Technology/CinemaCLIP)'s released encoder (its label schema 1.0.0) reads five lighting | |
| heads; `shot_light` takes, in order: silhouette; the key light's side (from the left, from the right, from above, backlit; left | |
| and right are the picture's; front and ambient name none); rim light (edge light, unless backlit); low key (high contrast, mean | |
| luminance under 0.35) or high key (low contrast, over 0.65). The head's `under` label names nothing: of the four faces among the | |
| first 100 read as lit from below, the three checked by eye on a sheet were frontal flash shots (the `raw/` lighting run still carries "from below" from the | |
| first rule). By eye on that sheet, high key (6) and low key (1) were right, and from above, from the left and from the right were | |
| right or generic (daylight from above). Counts over every face: from above 15,436, from the right 3,341, high key 2,381, rim light 1,353, low key 1,101, from the left 1,033, backlit 63; none 47,303. The reader names the right of the picture | |
| about three times as often as the left (3,341 against 1,033); the faces checked by eye were lit from the side it named, | |
| but the imbalance itself is unexplained and may be the reader's own. | |
| FFHQ holds people of every age, children and babies included; the captions give the apparent age as seen. | |
| Code: [github AbstractEyes/qwen-symbolic-thinking](https://github.com/AbstractEyes/qwen-symbolic-thinking) (`captions.run`, | |
| `lighting.heads_run` v0.1.49; `absences.drop` v0.1.52, the light rule v0.1.50; `notebooks/ffhq_captions.ipynb`). The faces were read from | |
| [gaunernst/ffhq-1024-wds](https://huggingface.co/datasets/gaunernst/ffhq-1024-wds), FFHQ's `images1024x1024` transcoded losslessly to | |
| WebP under the original numbers. | |
| ## Uses | |
| Portrait captioning and caption evaluation, tag prediction against a booru vocabulary, lighting-aware captioning, and conditioning | |
| research for image models. Not for the development or improvement of facial recognition technologies (FFHQ's terms), and nothing | |
| here names or identifies anyone. | |
| ## License | |
| CC BY-NC-SA 4.0, as FFHQ itself: non-commercial use, credit by citing the paper below, changes indicated, and works built on this | |
| shared under the same license. FFHQ's images are not here; each keeps its own Flickr license (Creative Commons BY 2.0, BY-NC 2.0, | |
| Public Domain Mark 1.0, CC0 1.0 or U.S. Government Works), which FFHQ's metadata names per image. Made with | |
| [Qwen3.5-4B](https://huggingface.co/Qwen/Qwen3.5-4B) (Apache-2.0), [animetimm/eva_giant_patch14_560.dbv4-full](https://huggingface.co/animetimm/eva_giant_patch14_560.dbv4-full) | |
| (GPL-3.0) and [CinemaCLIP](https://huggingface.co/OZU-Technology/CinemaCLIP) (OpenRAIL-M). | |
| ## Privacy | |
| FFHQ lets photo owners leave it (the `no_cv` tag on Flickr, a license change, removal, or a request to NVIDIA), and NVIDIA updates the | |
| set. The rows here are keyed by image number so that a face removed from FFHQ can be removed here too: open a discussion on this | |
| repository with the image number. | |
| ## Citation | |
| ```bibtex | |
| @article{karras2018stylegan, | |
| title = {A Style-Based Generator Architecture for Generative Adversarial Networks}, | |
| author = {Karras, Tero and Laine, Samuli and Aila, Timo}, | |
| journal = {arXiv preprint arXiv:1812.04948}, | |
| year = {2018} | |
| } | |
| ``` | |