CivArchive
    mosaic detection(All-in-one) - v3.0
    NSFW
    Preview 101793316
    Preview 101794134
    Preview 101794350
    Preview 102910571

    Yolo26 version is newest.

    The zip file contains two models of different sizes (S and L) and a simple workflow.

    Basically, S is sufficient, but you can use whichever you prefer.

    Mosaic Detection series is integrated. penis, pussy, and penetration can be detected simultaneously.

    This model is Adetailer for i2i or inpaint. GameCG,ArtistCG,and other.

    Detects a mosaic censorsip, and redraw. Use it in conjunction with Lora or prompt as you like.

    • Recommended for use with comfyui.

    • denoising strength: around 0.5 ⇒Adjust it so that you can draw it well.

    • It's important to write detailed prompts.

    Differences from previous version ↓

    1. Improved detection accuracy.

    2. If the angle is from the front where the stomach can be seen, the detection range will be wider. !BBOX version only!

    ⇒Possible to drawing pubic_hair's on-off,stomach_bulge,x-ray,uterus, etc...

    In that case,increase denoising strength.

    !!Atention!!

    This model was trained with yolo11, so don't forget to update ultralytics.

    \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

    モザイク検出シリーズは統合されました。ペニス、マンコ、挿入を同時に検出できます。

    このモデルはi2iまたはinpaint用のAdetailerです。ゲームCG、アーティストCGその他に。

    モザイク修正を検出して再描画します。好みのLoraやプロンプトと合わせて使ってください。

    Comfyuiでの使用を推奨

    denoising strength:0.5前後 ⇒うまく描画できるように調整してください。

    プロンプトは詳細に書いてください。

    以前のバージョンとの違い ↓

    1.検出精度の向上

    2.お腹が見える正面からのアングルの場合、検出範囲が広くなります。!BBOX versionのみ!

     ⇒陰毛の有無、腹ボコ、透視、子宮などの描画が可能に。

      その場合、denoising strengthを上げてください。

    !!注意!!

    このモデルはyolo11でトレーニングしたため、ultralyticsのアップデートを忘れずに。

    Description

    segmentation version UPDATED!

    FAQ

    Comments (16)

    bevas55694654Sep 22, 2025· 2 reactions
    CivitAI

    Will you update V3 in safetensor format?

    berezka_dudeSep 25, 2025
    CivitAI

    How can i use it with ComfyUI?

    wildcats
    Author
    Sep 25, 2025· 1 reaction

    see my posted image on v3.0

    2829838235556Sep 28, 2025
    CivitAI

    Hello, could you please upload a complete workflow? The workflow I built based on your diagram isn't working.

    wildcats
    Author
    Sep 28, 2025

    Please use the 4th image I posted.

    g1263495582Sep 29, 2025
    CivitAI

    thanks, can you make a censor bar detection version?

    wildcats
    Author
    Sep 29, 2025· 1 reaction

    A test version has already been created, and I'm currently adjusting it for the upgrade.Wait a minute.

    g1263495582Oct 4, 2025

    @wildcats From my testing, it's not better for no color, and some images aren't fully segmented. There are still mosaic pixels remaining in a few spots.

    wildcats
    Author
    Oct 4, 2025

    @g1263495582 Perfect detection is impossible, so please adjust by widening the mask area, lowering the threshold, etc.

    monsternova06880Oct 1, 2025· 2 reactions
    CivitAI

    Great model but is it possible to handle video? I tried to supply gif as input but the segm nodes returned:

    ERROR: DetailerForEach does not allow image batches.

    if this is not mean for video, i do really happy if community can suggest other mosaic remover for videos !!!

    wildcats
    Author
    Oct 3, 2025

    It only learns from still images, not videos.

    g1263495582Oct 3, 2025

    I think u need separate a video into each of its frames as an image sequence.

    wildcats
    Author
    Oct 3, 2025

    @g1263495582 It is possible to learn video using bbox, but I don't know if redraw is possible.

    wildcats
    Author
    Oct 5, 2025· 1 reaction

    I created a simple workflow to split a video into frames and detect mosaic.But when switch back to video, the redraw is inconsistent.

    10393407Dec 16, 2025

    @wildcats what if use T2V or I2V WAN or better WAN 2.2 FUN (https://huggingface.co/PineAmbassador/Wan2.2-VACE-Fun-A14B-NSFW_Merge?not-for-all-audiences=true) with
    1 mask as input that we got from you detector
    2 reference frame as one image from I2I (to get accurate image without mosaic)
    Looks promising. We can even make some more images into mix and model will make up transitions. At least I think so

    eloraKMar 5, 2026

    Dude you need to convert the batch into a list

    Detection
    Other

    Details

    Downloads
    441
    Platform
    CivitAI
    Platform Status
    Available
    Created
    9/22/2025
    Updated
    5/14/2026
    Deleted
    -

    Files

    mosaicDetectionAllIn_v30.pt

    Mirrors

    HuggingFace (1 mirrors)
    CivitAI (1 mirrors)