Model Training: LoRa

Close-up of Julie Newmar Catwoman portraying her face with mask and cat ears, background of egyptian cat statue in a history museum
Close-up of Julie Newmar Catwoman portraying her face with mask and cat ears
Medium shot of Julie Newmar Catwoman standing inside a history museum with egyptian cat statues in the background
Julie Newmar Catwoman crouching in action pose on a rooftop with a background of a city at night
Close-up of Julie Newmar Catwoman portraying her face with mask and cat ears

Project information

  • Category: Model Training

This project focuses on training a LoRA model for Julie Newmar's Catwoman, the iconic 1966 Batman character. LoRA, or Low-Rank Adaptation, enables efficient fine-tuning of diffusion models for specific concepts, like characters, with smaller file sizes and faster training times.

I gathered a diverse dataset of Julie Newmar Catwoman images, curated for quality and relevance. Captions describing pose, expression, and background were created for each image, serving as prompts for the LoRA model. Training was done on Stable Diffusion XL, optimizing settings and hyperparameters for quality.

The project aimed to explore diffusion models, refining data collection, curation, and training methods. Choosing a complex character challenged the model to achieve optimal accuracy.