CoolFace
Datasetpublic

4Fever4/siwa-kershef-architecture

Siwa Kershef Vernacular Architecture — curated captioned dataset 112 curated, hand-captioned photographs of the earthen (kershef: salt-mud and rock) architecture of Siwa Oasis, Egypt — Shali fortress, Aghurmi, restored houses, the old mosque, and new buildings built in the traditional technique. Built to train a style LoRA whose trigger k3rshef carries the material/massing vocabulary of the style without bleeding into other architectural styles. How it was built… See the full description on the dataset page: https://huggingface.co/datasets/4Fever4/siwa-kershef-architecture.

sourceHugging Facecc-by-sa-4.0updated 18h agoView on Hugging Face
0likes24downloads
Dataset Card

Siwa Kershef Vernacular Architecture — curated captioned dataset

112 curated, hand-captioned photographs of the earthen (kershef: salt-mud and rock) architecture of Siwa Oasis, Egypt — Shali fortress, Aghurmi, restored houses, the old mosque, and new buildings built in the traditional technique. Built to train a style LoRA whose trigger k3rshef carries the material/massing vocabulary of the style without bleeding into other architectural styles.

How it was built

StageCountMethod
Candidates1,398Wikimedia Commons categories: Architecture of Siwa, Shali (+subcats), Aghurmi, Adrere Amellal, Dovecote towers in Siwa, Siwa Town (+subcats). Only files ≥ 900 px short side with CC0 / PD / CC BY / CC BY-SA licences kept
Downloaded921licence + author + source page saved per file
Auto-filter896perceptual-hash near-duplicate removal (pHash distance ≤ 6) + CLIP ViT-L/14 relevance score (architecture vs people / food / landscape / sky / documents)
Manual curation112reviewed on contact sheets; rejected burst duplicates, skyline/sunset shots, markets & tourists, signage, B&W, the Pharaonic stone Oracle temple and modern stone buildings (not kershef)

Balance decision: Commons is dominated by photographs of the Shali ruins. Left unchecked, a LoRA learns "decay" as the style. Ruins were capped to ~45 % and every image is captioned with its condition (restored, partially restored, newly built, eroded ruin) so that condition is separable from style.

GroupImages
intact / restored / newly built exteriors46
interiors (tapered kershef columns, palm-trunk ceilings)15
Shali & Aghurmi ruins, details, openings51

Captions

Hand-written with a controlled vocabulary — see CAPTIONING.md. Format:

k3rshef, <view>, <subject>, <elements...>, <condition>, <context>, <light>, photo

Element words (palm-trunk beams, thick tapered columns, small window openings, external staircase, crenellated parapet …) are captioned so they can be requested or omitted at inference. Material, texture and colour are intentionally not captioned so they bind to the trigger.

Licensing & attribution

Each image keeps its original licence. train/metadata.csv lists, per image: source page, author, licence and the caption. Images: 57 Public Domain, 33 CC BY-SA 4.0, 10 CC BY 4.0, 7 CC BY 3.0, 5 CC BY 2.0. Captions and curation: CC BY-SA 4.0.

reg/ — regularisation images (used by the v2 experiment)

48 images generated with base SDXL (not real photos) of neighbouring styles — Nubian, modern Cairo, Mamluk, Ottoman, Mediterranean, riad, desert resort, etc. — captioned without the trigger, used for prior preservation in the v2 LoRA. See the model repo for why v2 over-regularised and v1 at strength 0.6 is the recommended result.