Using this LORA, you can obtain images of a beautiful character looking back from a rear perspective. It has been a while since I trained using standard methods, and the results look good, though I want to create some unusual characters. It seems that flexibility is not as great for male characters or animals.
このLORAを使用すると、美しいキャラクターが振り返る動作を背後からの視点で捉えた画像を得ることができます。長い間、正規の方法でのトレーニングは行っていませんでしたが、効果は悪くないようです。しかし、奇妙なキャラクターを作成したいと考えており、男性キャラクターや動物の場合、柔軟性がそれほど高くないようです。
Description
FAQ
Comments (3)
蠻有意思的意外產物是..我提詞某國總統,結果居然出現了兩個人,一個是總統本人一個是他老婆,畫面是他把他老婆做出回眸的動作的然後總統把她裙子掀起....,這東西很有意思但是發了怕我被抓走..
Lol what about telling us the prompt, with redacted names, instead of the pic?
@Honest_Abe The image depicts obama is posing on a table, kneeling in such a way that the dress obscures her lower back, . <lora:ptgs:0.8>,city,
Negative prompt: shose
Steps: 80, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 1107887153, Size: 1024x1024, Model hash: a928fee35b, Model: albedobaseXL_v20, VAE hash: 63aeecb90f, VAE: sdxl_vae.safetensors, Lora hashes: "ptgs: 46119fbabf5d", Downcast alphas_cumprod: True, Version: v1.8.0, Hashes: {"vae": "63aeecb90f", "lora:ptgs": "dbb4d5771f", "model": "a928fee35b"}