Use this LoRA to add a unique atmospheric & cinematic look to your image gens!

Art Series #1: Shot by Elsa Bleda (@elsableda). I wanted to try something outside my wheelhouse of celebrity LoRAs, and decided to create a style LoRA for a photographer whose work I admire. This LoRA is an homage to her work (which I highly recommended you check out) and only seeks to pay tribute, and not replace her work in any way, shape or form.

Check out my checkpoint merge here: AneurysmMix . Download my other LoRAs here

Feedback (comments/reviews/likes) are most welcome, as are tips/buzz so I can keep making these! I'd love to see the images you create, so please share on here or reach me on Discord (zombiehead91). I don't usually get too many shares on my celeb LoRAs so I'm hoping more people are able to share their image gens with this one.

About the subject:

A South Africa-based photographer and visual artist who defines herself on Instagram as “Lover of little details. Fan of east Asian cinema.” Her work explores various themes of isolation, illumination and landscapes, producing incredibly atmospheric images that retain a unique cinematic quality. Most of her work involves documenting the everyday world around her, with a particular interest in landscapes featuring human interventions, placing a strong emphasis on color, composition and lighting.

Trigger word: ZH_Shot_by_Elsa_Bleda,

Other useful triggers:

For scenery/landcapes: scenery, no humans, realistic, outdoors, wide shot, indoors, building, city, cityscape, skyscaper, ocean, water, beach, nature, forest, sunset, blue sky, cloudy sky, night sky, etc.

For human subjects included : 1girl, 1boy, multiple girls, multiple boys, solo, realistic,

Please see tags in LoRA for full list, you'll find many more useful tags in there.

About the sample images

Sample images generated with minimal prompting, using hi-res.fix but WITHOUT other LoRAs for demonstration of the LoRA's capabilities. ADetailer use is recommended.
Suggested settings : 0.6 to 1.0, Clip Skip: 1, ADetailer: yes
Method: DPM++SDE Karras, Steps:30-40, CFG : 6-7, 512x768, 768x512 + high res fix with 2x upscale

About the training data:

This LoRA was trained on a selected mix of images involving human subjects and a larger variety of pure landscape imagery.

Please do not share this LoRA anywhere else without my knowledge, or for any commercial use.