
An Improvement on the first model.
Added more images to dataset.
Better captions:
-
captioning included tags for year, bodystyle (hatchback, notchback, ghia)
-
also captioned headlights, grill, tail-lights, as well as general camera angle (view from front, view from behind, view from left, view from right)
use htk-mstng2 to trigger.
Usage:
-
Use notchback/hatchback to not get weird SUV/crossover.
-
Use at normal 1.0 strength to get normal images.
-
Reduce strength to use to get that 70's mustang /Ford look to other things. or to try to include other subjects. Use part prompts like grill, headlights, -
use with style/material LoRas to get odd variants. (the reason I created/uploaded this)
Experiment: added interior shows to the dataset. interior shots use the tag htk-mstg2int
not great, but makes an attempt to give interior. will likely explore this concept further.
... working on an interior LoRa because:
-
a comment made me want to do it.
-
It is hard.
Overall I am happy with this one. Is shows some versatility, even attempts to be in a image with other vehicles, and works with other LoRas like hovering car, which is one of the reasons I made this.
描述:
Same dataset, this time fluxed
Has a weird bias for showing the back
训练词语: htk-mstng2,hatchback,notchback,view from front
名称: ht-mstng2-FLUX_1000.safetensors
大小 (KB): 167939
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success