ComfyUI Node
StereoDiffusion
A ComfyUI node in image/stereo with 17 inputs and 3 outputs.
StereoDiffusion
- image
- depth_map
- model
- clip
- vae
- stereo_pair
- left_image
- right_image
◄scale_factor5.0►
◄directionuni►
◄deblurfalse►
◄pipeline_modeFast (Warp + Inpaint)►
◄guidance_scale3.0►
◄num_inference_steps20►
◄seed1337►
◄null_text_optimizationtrue►
◄denoise_strength0.60►
◄model_idrunwayml/stable-diffusion-v1-5►
◄inpaint_model_idrunwayml/stable-diffusion-inpainting►
◄prompt►
Categoryimage/stereo
Inputs (17)
| Name | Type | Default | Description |
|---|---|---|---|
| image | IMAGE | — | |
| depth_map | IMAGE | — | |
| scale_factor | FLOAT | 5.01–20 | Controls the strength of the stereo effect (disparity) |
| direction | COMBO | uni | uni: unidirectional attention, bi: bidirectional attention |
| deblur | BOOLEAN | false | Add noise to unfilled regions to avoid blurring |
| pipeline_mode | COMBO | Fast (Warp + Inpaint) | Standard: DDIM inversion (high quality, slow). Fast: Warp image with depth, then AI-inpaint only the gap regions (fast, works with turbo/LCM models) |
| guidance_scale | FLOAT | 3.00–20 | CFG scale. Standard mode: 3-10. Turbo models: 0.0. LCM: 1.0-2.0 |
| num_inference_steps | INT | 201–100 | Number of inference steps. Standard DDIM: 30-100 (default 50). Fast inpainting: 20-30. Turbo/LCM: 1-8 |
| seed | INT | 13370–18446744073709550000 | Random seed for reproducible results |
| null_text_optimizationopt | BOOLEAN | true | Enable null-text optimization for better reconstruction (Standard mode only) |
| denoise_strengthopt | FLOAT | 0.600.1–1 | How much noise to add before denoising (Fast mode). Lower = preserve original more, Higher = more model creativity for filling gaps |
| modelopt | MODEL | ComfyUI MODEL input. Fast mode: use inpainting model (9ch UNet). Standard mode: use any SD1/SD2 model. | |
| clipopt | CLIP | CLIP from Load Checkpoint | |
| vaeopt | VAE | VAE from Load Checkpoint | |
| model_idopt | STRING | runwayml/stable-diffusion-v1-5 | Fallback HuggingFace model ID (Standard mode, when no ComfyUI model connected) |
| inpaint_model_idopt | STRING | runwayml/stable-diffusion-inpainting | Fallback inpainting model ID (Fast mode, when no ComfyUI model connected) |
| promptopt | STRING | Optional text prompt to guide inpainting (Fast mode). Describe the image content for better gap filling. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| stereo_pair | IMAGE | — |
| left_image | IMAGE | — |
| right_image | IMAGE | — |