sketchkit.sketch2image.models package

Submodules

sketchkit.sketch2image.models.controlnet module

ControlNet Model Implementation for Sketch-to-Image Conversion

class sketchkit.sketch2image.models.controlnet.ControlNetModel(device: str = 'cuda', controlnet_model: str = 'lllyasviel/sd-controlnet-scribble', sd_model: str = 'runwayml/stable-diffusion-v1-5', torch_dtype: dtype = torch.float16, safety_checker: bool = False, **kwargs: Any)

Bases: object

ControlNet model for scribble-conditioned sketch-to-image conversion.

generate(control_image: Image, prompt: str, size: int | Tuple[int, int] = 512, negative_prompt: str = '', num_inference_steps: int = 20, guidance_scale: float = 7.5, controlnet_conditioning_scale: float = 1.0, seed: int | None = None, **kwargs: Any) → Image

Generate image from scribble-conditioned sketch.

sketchkit.sketch2image.models.pix2pix module

Pix2Pix Model Implementation for Sketch-to-Image Conversion

class sketchkit.sketch2image.models.pix2pix.Pix2PixModel(device: str = 'cuda', model_name: str = 'timbrooks/instruct-pix2pix', torch_dtype: dtype = torch.float16, safety_checker: bool = False, **kwargs: Any)

Bases: object

InstructPix2Pix model for direct sketch-to-image translation.

generate(control_image: Image, prompt: str, size: int | Tuple[int, int] = 512, negative_prompt: str = '', num_inference_steps: int = 10, guidance_scale: float = 7.5, image_guidance_scale: float = 1.0, seed: int | None = None, **kwargs: Any) → Image

Generate image from sketch using instruction-based translation.

sketchkit.sketch2image.models.t2i_adapter module

T2I-Adapter Model Implementation for Sketch-to-Image Conversion

class sketchkit.sketch2image.models.t2i_adapter.T2IAdapterModel(device: str = 'cuda', adapter_model: str = 'TencentARC/t2iadapter_canny_sd15v2', sd_model: str = 'runwayml/stable-diffusion-v1-5', torch_dtype: dtype = torch.float16, safety_checker: bool = False, **kwargs: Any)

Bases: object

T2I-Adapter model for edge-conditioned sketch-to-image conversion.

generate(control_image: Image, prompt: str, size: int | Tuple[int, int] = 512, negative_prompt: str = '', num_inference_steps: int = 50, guidance_scale: float = 7.5, adapter_conditioning_scale: float = 1.0, seed: int | None = None, **kwargs: Any) → Image

Generate image from edge-conditioned sketch.

Module contents

Sketch2Image Models Module

This module contains all AI model implementations for sketch-to-image conversion.

class sketchkit.sketch2image.models.ControlNetModel(device: str = 'cuda', controlnet_model: str = 'lllyasviel/sd-controlnet-scribble', sd_model: str = 'runwayml/stable-diffusion-v1-5', torch_dtype: dtype = torch.float16, safety_checker: bool = False, **kwargs: Any)

Bases: object

ControlNet model for scribble-conditioned sketch-to-image conversion.

generate(control_image: Image, prompt: str, size: int | Tuple[int, int] = 512, negative_prompt: str = '', num_inference_steps: int = 20, guidance_scale: float = 7.5, controlnet_conditioning_scale: float = 1.0, seed: int | None = None, **kwargs: Any) → Image

Generate image from scribble-conditioned sketch.

class sketchkit.sketch2image.models.Pix2PixModel(device: str = 'cuda', model_name: str = 'timbrooks/instruct-pix2pix', torch_dtype: dtype = torch.float16, safety_checker: bool = False, **kwargs: Any)

Bases: object

InstructPix2Pix model for direct sketch-to-image translation.

generate(control_image: Image, prompt: str, size: int | Tuple[int, int] = 512, negative_prompt: str = '', num_inference_steps: int = 10, guidance_scale: float = 7.5, image_guidance_scale: float = 1.0, seed: int | None = None, **kwargs: Any) → Image

Generate image from sketch using instruction-based translation.

class sketchkit.sketch2image.models.T2IAdapterModel(device: str = 'cuda', adapter_model: str = 'TencentARC/t2iadapter_canny_sd15v2', sd_model: str = 'runwayml/stable-diffusion-v1-5', torch_dtype: dtype = torch.float16, safety_checker: bool = False, **kwargs: Any)

Bases: object

T2I-Adapter model for edge-conditioned sketch-to-image conversion.

generate(control_image: Image, prompt: str, size: int | Tuple[int, int] = 512, negative_prompt: str = '', num_inference_steps: int = 50, guidance_scale: float = 7.5, adapter_conditioning_scale: float = 1.0, seed: int | None = None, **kwargs: Any) → Image

Generate image from edge-conditioned sketch.