PennyLucent

I'm training myself to generate random character, generate 100 images from the first image with Qwen, then train a LORA.

For this character I used this prompt

score_9, score_8_up, score_7_up, Stable_Yogis_PDXL_Positives, real life photo, perfect eyes, detailed face, ((Masterpiece, Best_Quality, Highest_Resolution)), amateur, flash, 25 years old, pretty girl, 1girl, realistic, photorealistic, half-body portrait, oval face shape, soft jawline, rounded chin, high cheekbones with subtle rosy flush, straight refined nose with narrow bridge and discreet nostrils, small mouth with subtle smile, full lips (thin upper lip, plump lower lip forming gentle cupid's bow), arched thin eyebrows with natural expressiveness, large almond-shaped eyes slightly upturned at outer corners, long dark lashes, warm hazel irises gazing sideways, smooth tanned skin with faint freckles, wavy shoulder-length brown hair framing face, collarbone visible, generous firm breasts (medium to large size, natural D-cup volume), deep plunging cleavage, black ribbed bikini top hugging curves, taut fabric over rounded contours, subtle natural bounce, exposed upper chest with sun-kissed glow, navel partially visible, natural lighting, rating:explicit, lips, detailed skin texture, breasts, topless

Then I use this kind of prompt to manage with Qwen :

positive: looking away, smiling, eyes closed, sitting, portrait, straight-on, tank top, bedroom, soft light
negative: looking at viewer, angry, eyes open, standing up, full body, from below, bikini, outdoor beach, harsh light
----
positive: looking back, laughing, mouth open, standing up, upper body, three quarter, sundress, park/garden, golden hour/sunset light
negative: looking at viewer, crying, mouth closed, laying down, lower body, from above, formal dress/gown, indoor kitchen, dark light

And after that I review image by hands and launch training on Civitai !