New Dataset and captioning with Florence2





Inspired and influenced by mnemic's original SD1.5 LoRA. This is an updated version 3 of my previous two releases for Flux.1 Dev. I used a fresh dataset generated using the original LoRA and revAnimated Checkpoint. Captioning was done with Florence2
This LoRA will add a 1950s spy theme to your images, desaturated/muted colors and "hidden" cameras. Gadgets and technology can take on a look of cameras, lenses and related.
Use the trigger word: "YFG-SpyWorld50s" along with the LoRA to activate it.
Example prompt: YFG-SpyWorld50s 1950s microwave over in a retro kitchen.
0.6-0.8 LoRA strength works well but anywhere from 0.25 to 1.10 strength should be explored. Trigger word is not necessary as this model was trained to be used with or without the trigger. Short or long prompts work well. Flux guidance can have tremendous effect from 2.0 to 6.0 and worth exploring in between for interesting results.
Don't pass up the opportunity to mix with other LoRAs. The model is very flexible and I have had success using it together with Eldritch Adam's Digital Painting LoRA. Use at a mix of .65% for each LoRA.
The following words may enhance the level of spying-ness and gadgetry
spy-theme
50s
secret tech
wires
spy-craft
cameras
spyPlease post your images below and feedback is appreciated. Also post any findings on how you have used the model in different ways.
Thank you.
New Dataset and captioning with Florence2
1. 轉载模型僅供學習與交流分享,其版權及最終解释權归原作者。
2. 模型原作者如需認領模型,請通過官方渠道联系海藝AI工作人員進行認證。我們致力於保護每一位創作者的權益。 點擊去認領
