Skip to main content

SD15DepthControlNetModel

GenerativeModel
DashAI.back.models.hugging_face.SD15DepthControlNetModel

Depth-conditioned ControlNet pipeline built on Stable Diffusion 1.5.

Takes an input image and a text prompt. A depth map is estimated from the image using Intel's DPT-Hybrid-MiDaS model, then fed as a spatial conditioning signal into the lllyasviel/sd-controlnet-depth ControlNet backbone together with the runwayml/stable-diffusion-v1-5 diffusion pipeline. The result is a 512 x 512 image that respects both the text description and the 3-D structure of the original scene.

References​

Parameters​

num_inference_steps : integer, default=20
Number of denoising steps to run. More steps refine the image but increase generation time. Typical range: 20-30 for fast results, 40-50 for higher quality.
controlnet_conditioning_scale : number, default=1.0
Weight of the ControlNet depth conditioning relative to the base diffusion pipeline (range 0.0-2.0). At 0.0 the depth map has no effect; at 1.0 the output closely follows the input structure; above 1.5 depth dominates and may produce rigid results.
guidance_scale : number, default=7.5
Classifier-Free Guidance (CFG) scale. Controls how strictly the image follows the text prompt. Values 7-9 are typical for SD 1.5.
device : string, default=CPU
Hardware device for inference. GPU is strongly recommended for diffusion models. CPU inference is possible but very slow.

Methods​

generate(self, input: Tuple[ForwardRef('Image.Image'), str]) -> List[Any]

Defined on SD15DepthControlNetModel

Generate output from a generative model.

Parameters

input : Tuple[Image.Image, str]
Input image and text prompt.

Returns

List[Any]
Generated output images in a list.

get_schema(cls) -> dict

Defined on ConfigObject

Generates the component related Json Schema.

Returns

dict
Dictionary representing the Json Schema of the component.

validate_and_transform(self, raw_data: dict) -> dict

Defined on ConfigObject

It takes the data given by the user to initialize the model and returns it with all the objects that the model needs to work.

Parameters

raw_data : dict
A dictionary with the data provided by the user to initialize the model.

Returns

dict
A validated dictionary with the necessary objects.

Compatible with​