Technicolor-50s-Diffusion
v2
terrariyum
Style Description
- highly-saturated postcard-like colors, flat high-key lighting, strong rim-lighting, 40s and 50s lifestyle
Recommended Usage
- Your prompt must include "tchnclr style"
- Use CFG of 7 or 8 for best results
- The model was trained with and excels at closeup portraits of men and women
- Try including "glitter" in your prompt!
- Putting "b&w" as a negative prompt will help ensure color image
Known Limitations
- It strongly tries to insert 40s and 50s hairstyles, clothing, and scenery
- As you can see from the examples, you can insert some modernity and blend with other styles. But if your prompt insists on all modern elements, the technicolor effect may disappear.
- The model tends to turn men into women. It also likes to add hats!
Training Process
20 images from movies filmed in technicolor, 200 photo-like classifiers, 6000 steps, using the Dreambooth Extension for Automatic1111.
What is Technicolor-50s-Diffusion?
Technicolor-50s-Diffusion is a highly specialized Image generation AI Model of type Safetensors / Checkpoint AI Model created by AI community user terrariyum. Derived from the powerful Stable Diffusion (SD 1.5) model, Technicolor-50s-Diffusion has undergone an extensive fine-tuning process, leveraging the power of a dataset consisting of images generated by other AI models or user-contributed data. This fine-tuning process ensures that Technicolor-50s-Diffusion is capable of generating images that are highly relevant to the specific use-cases it was designed for, such as movie, film, 1940s.
With a rating of 4.5 and over 2 ratings, Technicolor-50s-Diffusion is a popular choice among users for generating high-quality images from text prompts.
Can I download Technicolor-50s-Diffusion?
Yes! You can download the latest version of Technicolor-50s-Diffusion from here.
How to use Technicolor-50s-Diffusion?
To use Technicolor-50s-Diffusion, download the model checkpoint file and set up an UI for running Stable Diffusion models (for example, AUTOMATIC1111). Then, provide the model with a detailed text prompt to generate an image. Experiment with different prompts and settings to achieve the desired results. If this sounds a bit complicated, check out our initial guide to Stable Diffusion – it might be of help. And if you really want to dive deep into AI image generation and understand how set up AUTOMATIC1111 to use Safetensors / Checkpoint AI Models like Technicolor-50s-Diffusion, check out our crash course in AI image generation.
Popularity
Info
Latest version (v2): 1 File
About this version: v2
- higher quality output
- better match to the technicolor style
- greater flexibility for diverse prompts
- not strictly 1950's imagery
- Add "tchnclr" and/or "1950s" to enhance the technicolor effect
- The model was trained with and excels at closeup portraits of men and women
- The model makes beautiful hair, and describing the hair style helps
- Generating at 1024x with hires fix produces much better results than 512
- Use CFG of 4-7 for best results. Yes, 4!
- Faces tend to look similar. To avoid this, prompt a celebrity at low weight or use an alternating prompt, e.g "[woman|celebrity-name]"
- The more modernity that the prompt contains modern elements, the weaker the technicolor effect
- Blending with other artists and art styles reduces the technicolor effect but is worth trying
- d8ahazard's Dreambooth Extension for Automatic1111 webUI
- 40 high-quality training images from movies filmed in technicolor, manually captioned
- a version with 600 photo-like classifiers merged with a version with no prior preservation to balance style and flexibility
- 12,000 steps with a 2e-7 learning rate
