Model
Trends
.ai
Model
Trends
.ai
model AI sumber terbuka terbaik

Nod Grow Mask, Feather Mask and Invert Mask dalam ComfyUI

The core mask tools expand (Grow Mask), soften (Feather Mask) and flip (Invert Mask) a MASK so inpainted regions blend into the surrounding picture without visible seams.
Pek nod: ComfyUI core
Kategori: Topeng dan inpainting
Halaman ini belum diterjemahkan, jadi ia dipaparkan dalam bahasa Inggeris.

Apa yang dilakukan Grow Mask, Feather Mask and Invert Mask

A mask painted by hand or produced by a detector is usually too tight and too hard. Grow Mask dilates it by a number of pixels (negative values shrink it) and can round the corners with tapered_corners. Feather Mask fades the edges over left, top, right and bottom distances, turning a hard border into a gradient so the denoised pixels blend into the originals. Invert Mask swaps 0 and 1 to protect what was selected and edit everything else.
The usual chain for inpainting is Load Image -> MASK -> Grow Mask (8-20 px) -> Feather Mask -> Set Latent Noise Mask. For compositing with Image Composite Masked, feathering alone is enough. The Impact Pack and other suites add blur, threshold and combine operations, but these three core nodes cover most work.

Input

Nama
Jenis
Apakah ia
mask
MASK
The mask to modify.
expand (Grow Mask)
INT
Pixels to grow (positive) or shrink (negative).
tapered_corners (Grow Mask)
BOOLEAN
Rounded growth instead of square.
left/top/right/bottom (Feather Mask)
INT
Fade distance on each side in pixels.

Output

Nama
Jenis
Apakah ia
MASK
MASK
The modified mask.
AD
Langkau persediaan: aliran kerja ComfyUI prapasang di awan
BitVector menjalankan aliran kerja ComfyUI sedia ada pada GPU-nya sendiri. Tiada pemasangan, tiada nod yang hilang, tiada kotak merah. Buka di telefon atau komputer riba anda dan jana dalam seminit.

Cara menggunakan Grow Mask, Feather Mask and Invert Mask

  1. Start from a MASK output (Load Image, Load Image (as Mask) or a detector).
  2. Add Grow Mask with expand 8-20 so the new content overlaps the old slightly.
  3. Add Feather Mask with 8-24 on each side for a soft transition.
  4. Connect the result to the inpainting node or to Image Composite Masked.
  5. Preview the mask (Convert Mask to Image -> Preview Image) to check the shape.

Tetapan dan petua

  • Grow before you feather; feathering a tight mask leaves the seam inside the object.
  • Negative expand values clean up speckles from automatic masks.
  • For hair and fur, feather generously and keep denoise around 0.5-0.7.
  • Masks are float tensors; nothing stops you from multiplying two masks with Mask Composite for an intersection.

Penyelesaian masalah Grow Mask, Feather Mask and Invert Mask

A soft halo appears around the inpainted object

Mengapa ia berlaku
Too much feathering with a very high denoise: the gradient region is regenerated differently from both sides.

Cara membaikinya
Reduce feather to 8-12 px or lower denoise to 0.6-0.8.

Grow Mask changes nothing

Mengapa ia berlaku
The mask is already saturated (all 1) or the expand value is 0.

Cara membaikinya
Check the mask with a preview; set expand to a visible value such as 16.

Edits leak outside the painted area

Mengapa ia berlaku
The grown mask now covers neighbouring details.

Cara membaikinya
Reduce expand, or feather instead of growing.

AD
Jalankan aliran kerja ini dari telefon anda
Setiap aliran kerja di halaman ini telah prapasang di BitVector dengan model dan nod tersuai sedia ada. Pilih satu, taip prompt, selesai. Tiada persediaan, tiada apa untuk dimuat turun.

Soalan tentang Grow Mask, Feather Mask and Invert Mask

Is there a Blur Mask node in core?

Feather Mask covers the edge fade. For a full Gaussian blur use the Impact Pack or WAS suite blur nodes.

What does tapered_corners do?

It grows the mask with a rounded (cross-shaped) kernel instead of a square one, so corners stay rounded.

Nod berkaitan

Load Image (as Mask)
ComfyUI core
Load Image (as Mask) turns one channel of a picture (alpha, red, green or blue) into a MASK, so black-and-white mask files or painted layers can drive inpainting and compositing.
Set Latent Noise Mask
ComfyUI core
Set Latent Noise Mask attaches a MASK to a LATENT so the sampler only changes the white area: inpainting and regional regeneration with any normal checkpoint.
VAE Encode and VAE Encode (for Inpainting)
ComfyUI core
VAE Encode converts a pixel IMAGE into a LATENT for img2img; VAE Encode (for Inpainting) does the same while blanking the masked area so an inpainting model can repaint it.
FaceDetailer (Impact Pack)
ComfyUI-Impact-Pack
FaceDetailer finds faces (or hands, eyes) with a YOLO detector, crops each one, re-renders it at full resolution with your model and prompt, and pastes it back, fixing the small, mushy faces in wide shots.
AD
Penat membaiki nod? Biarkan awan melakukannya
BitVector mengekalkan beratus-ratus aliran kerja ComfyUI yang dipasang, dikemas kini dan diuji pada GPU awan yang pantas. Tiada Python, tiada ralat CUDA, tiada had VRAM. Berfungsi dari mana-mana pelayar.

Lagi nod ComfyUI

CLIP Text Encode (Prompt)
ComfyUI core
CLIP Text Encode turns a text prompt into CONDITIONING using the model text encoder. One node holds the positive prompt, a second one the negative prompt.
ComfyUI Manager
ComfyUI-Manager
ComfyUI Manager is the extension that installs, updates and fixes custom node packs and models from inside the interface, resolves missing nodes in imported workflows and snapshots your setup.
KSampler
ComfyUI core
KSampler runs the denoising loop: it takes the model, prompts and a latent and produces the finished latent image, controlled by seed, steps, cfg, sampler, scheduler and denoise.
Load Checkpoint
ComfyUI core
Load Checkpoint (CheckpointLoaderSimple) opens a .safetensors or .ckpt model file and hands out the three parts every workflow needs: the diffusion MODEL, the CLIP text encoder and the VAE.
Save Image
ComfyUI core
Save Image writes the IMAGE tensor to ComfyUI/output as a PNG, with the whole workflow embedded in the file metadata so the picture can be dragged back into ComfyUI to restore the graph.
VAE Decode
ComfyUI core
VAE Decode converts the sampled LATENT into a pixel IMAGE with the VAE; VAE Decode (Tiled) does the same in tiles for very large images.