Extensions/ComfyUI Speech Bubble Detector
ComfyUI Extension

ComfyUI Speech Bubble Detector

Speech bubble detector for ComfyUI based on SAM3 Text Segmentation with crop-locked region detection for manga/manhwa cleanup workflows.

By daminik124124-ops·Created 2 months ago·Updated 2 months ago· 0
daminik124124-ops/ComfyUI-Speech-Bubble-Detector
Nodes15
On cloudLocal install
CategorySAM3, SAM3/prompts
Stars0
Updated2 months ago

Nodes (15)

(Down)Load SAM3 Model

Get Meta's SAM3 weights into your ComfyUI graph

SAM3
SAM3 BBox Collector

Draw boxes on the image to guide SAM3

SAM3
SAM3 Combine Boxes

Merge single boxes into one prompt SAM3 accepts

SAM3/prompts
SAM3 Combine Points

Merge single points into one prompt SAM3 accepts

SAM3/prompts
SAM3 Create Box

Build a box prompt from center and size

SAM3/prompts
SAM3 Create Point

Build a single point prompt from numbers

SAM3/prompts
Speech Bubble Detector

Segment anything by typing what you want

Speech Bubble/SAM3
SAM3 Interactive Collector

SAM3's interactive collector

SAM3
SAM3 Multiprompt Segmentation

Mask several regions in one pass

SAM3
SAM3 Multi-Region Collector

Mark several objects for one multiprompt pass

SAM3
SAM3 Point Collector

Click your points right on the image

SAM3
SAM3 Propagate

Track your masked object across every frame

SAM3/video
SAM3 Point Segmentation

Point-and-box masking when text isn't precise enough

SAM3
SAM3 Video Output

Turn the tracked state into masks and frames you can use

SAM3/video
SAM3 Video Segmentation

Mark your target on the first frame

SAM3/video
Readme

ComfyUI Speech Bubble Detector

A ComfyUI custom node package for manga/manhwa speech bubble cleanup workflows.

This package is based on comfyui-sam3 and includes a patched SAM3 Text Segmentation workflow for crop-locked speech bubble detection.

Main node

Search in ComfyUI:

Speech Bubble Detector

This is the renamed/patched version of SAM3 Text Segmentation.

What is changed

  • The text segmentation node can use positive_boxes as crop-locked regions.
  • When positive_boxes are connected, detection runs only inside those regions and pastes masks back to the full image.
  • Added region_padding input for expanding each crop region.
  • Includes the attention fallback patch to reduce Flash Attention mask-warning spam.

Typical workflow

Use this together with the separate helper node:

JSON Regions to SAM3 Boxes API -> positive_boxes -> Speech Bubble Detector

Recommended settings:

text_prompt = speech bubble
confidence_threshold = 0.1 - 0.2
max_detections = -1
region_padding = 0 - 20

Install

Clone into ComfyUI/custom_nodes:

cd ComfyUI/custom_nodes
git clone https://github.com/daminik124124-ops/ComfyUI-Speech-Bubble-Detector.git

Then restart ComfyUI.

Important

Do not install this fork side-by-side with the original comfyui-sam3 unless you know what you are doing. This package is intended to replace the original SAM3 custom node package for this speech-bubble workflow.

The helper node ComfyUI-SAM3-JSON-Boxes-API remains a separate repository/package.

Credits

Based on the original ComfyUI-SAM3 project by PozzettiAndrea. This fork adds speech-bubble/crop-lock workflow changes.