This project is Lora to bring the Stable Diffusion model closer to the Japanese landscape.
The items I see in Japanese towns are Lora
Currently stores vending machines and mailboxes
Please use Lora with a numerical value of about 0.7 to 1
If adding elements to the image weakens Lora's expression, raise Lora's number to ~1 or add tags for the elements included in the data to reinforce it
The call tag and additional tag for each data are as follows
jihanki Lora(locon) :vending machine
Invocation tag: jihanki
Additional tags:
trash can, bicycle, building
jpost Lora: mailbox/post box
Invocation tag: jpost
Additional tags:
road, sign, bicycle, shop
日本の街で見かけるアイテムをLoraにしています
現在、自動販売機、ポストを格納しています
Loraの数値は0.7~1程度で使用して下さい
画像に要素を追加することでLoraの表現が弱くなる場合、Loraの数値を~1に上げるかデータに含まれる要素のタグを加えて補強してください
各データの呼び出しタグと追加タグは以下のとおりです
jihanki Lora(locon):自動販売機
呼び出しタグ:jihanki
追加タグ:
trash can,bicycle,building
jpost Lora:ポスト
呼び出しタグ:jpost
追加タグ:
road,sign,bicycle,shop