metadata
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
- 3D
- Cinematography
- OBJ
widget:
- text: >-
Isometric 3D Cinematography, a small boy stands in front of a playmat. The
playmat is adorned with a variety of toys, including a car, a fire
hydrant, and a fireman. The boy is wearing a short-sleeved blue t-shirt,
and tan pants. The floor is made of light-colored wood planks, and the
walls are painted white. To the right of the boy is a wooden shelf, and to
the left of the shelf is a green bench. The shelf is filled with various
toys, and there is a window in the background.
output:
url: images/IS1.png
- text: >-
Isometric 3D Cinematography, An outdoor view of a white dog with a black
collar. The dog is facing towards the right side of the image. There is a
white building to the right of the dog. There are windows on the building.
To the left of the white dog is a tree with yellow leaves on it. The tree
is casting a shadow on the ground. The ground is covered with leaves.
output:
url: images/IS2.png
- text: >-
Isometric 3D Cinematography, An aerial view of a large tree in the middle
of a road. The tree is surrounded by small bushes and shrubs. There are
two cars parked on the left side of the road. There is a small white house
on the right side. The house has a brown roof and white windows. The trees
are casting shadows on the ground.
output:
url: images/IS3.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Isometric 3D Cinematography
license: creativeml-openrail-m
The model is still in the training phase. This is not the final version and may contain artifacts and perform poorly in some cases.
Model description
strangerzonehf/Flux-Isometric-3D-Cinematography
Image Processing Parameters
Parameter | Value | Parameter | Value |
---|---|---|---|
LR Scheduler | constant | Noise Offset | 0.03 |
Optimizer | AdamW | Multires Noise Discount | 0.1 |
Network Dim | 64 | Multires Noise Iterations | 10 |
Network Alpha | 32 | Repeat & Steps | 28 & 3900 |
Epoch | 25 | Save Every N Epochs | 1 |
Labeling: florence2-en(natural language & English)
Total Images Used for Training : 24
Best Dimensions
- 768 x 1024 (Best)
- 1024 x 1024 (Default)
Setting Up
import torch
from pipelines import DiffusionPipeline
base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
lora_repo = "strangerzonehf/Flux-Isometric-3D-Cinematography"
trigger_word = "Isometric 3D Cinematography"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
Trigger words
You should use Isometric 3D Cinematography
to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.