zf-kbot/image-object-remover 🔢📝🖼️✓ → 🖼️
About
Example Output
Prompt:
"Identify the object marked by the grey mask in the image. Remove it completely and fill the area with a background that matches the surrounding environment. Ensure the lighting, texture, and perspective are consistent, leaving no trace of the removed object. Maintain the overall composition and quality of the image. Do not alter any other parts of the image."
Output
Performance Metrics
4.02s
Prediction Time
4.29s
Total Time
All Input Parameters
{
"seed": -1,
"steps": 4,
"prompt": "Identify the object marked by the grey mask in the image. Remove it completely and fill the area with a background that matches the surrounding environment. Ensure the lighting, texture, and perspective are consistent, leaving no trace of the removed object. Maintain the overall composition and quality of the image. Do not alter any other parts of the image.",
"quality": 100,
"batch_size": 1,
"output_ext": ".webp",
"input_image": "https://replicate.delivery/pbxt/OJjKW9MbHp4dmEoop7gqghOgACOhIouTQShp9MmbfonKPGCk/1bd2ba68-0c05-442d-9c0e-3421ac5f9836_original.jpg",
"use_mask_image": true,
"negative_prompt": "lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry",
"input_mask_image": "https://replicate.delivery/pbxt/OJjKVq2fxu0eEYYDIMUZqmaci7V19Weoq0a30jUdrhO2WigE/1bd2ba68-0c05-442d-9c0e-3421ac5f9836_mask.jpg"
}
Input Parameters
- seed
- 随机种子，-1 表示随机
- steps
- 去噪步数（steps）
- prompt
- 正向提示词（positive_prompt）
- quality
- 输出图片质量 (1-100)，对应 Save Image Extended 的 `quality`
- batch_size
- batch_size，决定一次生成的图片数量
- output_ext
- 输出图片后缀，例如 `.jpg` / `.png`
- input_image (required)
- 主输入图像（image 1）
- use_mask_image
- 是否使用遮罩图像
- negative_prompt
- 负面提示词（negative_prompt）
- input_mask_image (required)
- 遮罩图像（image 2）
Output Schema
Output
Example Execution Logs
Global interrupt (no prompt_id specified)
Checking inputs
✅ /tmp/inputs/2c07a20f-5b94-48e5-9071-2ad7c7be5b20.jpg
❌ /tmp/inputs/.webp not provided
✅ /tmp/inputs/0476488c-9466-4be8-a154-feeba412ce1d.jpg
====================================
Running workflow
got prompt
Executing node 201, title: Load Image, class type: LoadImage
Executing node 233, title: Get Image Size, class type: GetImageSize
Executing node 235, title: Math Calculate, class type: math_calculate
Executing node 236, title: Math Calculate, class type: math_calculate
Executing node 237, title: Show Any, class type: easy showAnything
Executing node 238, title: If else, class type: easy ifElse
Executing node 184, title: Load Image, class type: LoadImage
Executing node 214, title: Get Image Size, class type: GetImageSize
Executing node 223, title: Math Calculate, class type: math_calculate
Executing node 225, title: Math Calculate, class type: math_calculate
Executing node 220, title: If else, class type: easy ifElse
Executing node 202, title: Resize Images by Shorter Edge, class type: ResizeImagesByShorterEdge
Executing node 238, title: If else, class type: easy ifElse
Executing node 203, title: Convert Image to Mask, class type: ImageToMask
Executing node 197, title: Resize Images by Shorter Edge, class type: ResizeImagesByShorterEdge
Executing node 220, title: If else, class type: easy ifElse
Executing node 164, title: Image Crop By Mask And Resize (Swwan), class type: ImageCropByMaskAndResize
Executing node 173, title: Image Resize Sum, class type: Image_Resize_sum
Executing node 212, title: Draw Mask On Image (Swwan), class type: DrawMaskOnImage
🖼️ 正在解码潜空间/Decoding latent space...
✅ 解码完成/Decoding completed
Executing node 242, title: 使用涂抹图片开关, class type: easy imageSwitch
Executing node 144, title: 🐋 Qwen Image Integrated KSampler - Github:﹫luguoli, class type: QwenImageIntegratedKSampler
DrawMaskOnImage batch: 0%| | 0/1 [00:00<?, ?it/s]
DrawMaskOnImage batch: 100%|██████████| 1/1 [00:00<00:00, 363.46it/s]
Requested to load QwenImage
🎯 开始执行采样任务/Starting sampling task ......
🎲 种子/Seed: 6670266
📊 步骤数/Steps: 4
🎛️ CFG强度/CFG Scale: 1.0
🔄 降噪强度/Denoise: 1.0
🌀 采样器/Sampler: euler
📈 调度器/Scheduler: simple
📏 分辨率/Resolution: 0 x 0
🎯 生成模式/Generation Mode: 图生图 image-to-image
🔢 批量大小/Batch Size: 1
✨ AuraFlow Shift: 3.0
🎛️ CFG规范化强度/CFG Normalization Strength: 1.0
🧠 正向提示词长度/Positive Prompt Length: 360
🚫 负向提示词长度/Negative Prompt Length: 196
📝 指令长度/Instruction Length: 304
🗂️ 自动保存文件夹/Auto Save Folder: Disabled
📄 输出文件名前缀/Output Filename Prefix: auto_save
🧹 清理GPU内存/Clean GPU Memory: False
🗑️ 结束后清理CPU内存/Clean CPU Memory After Finish: False
🔊 声音通知/Sound Notification: False
✨ 应用shift参数/Applying shift parameter: 3.0
✅ shift参数已应用成功/Shift parameter applied successfully
🎛️ 应用强度/Applying strength: 1.0
✅ 规范化已应用成功/Normalization applied successfully
⏳ [Prompt] 正在生成正向条件 / Generating positive condition...
⏳ [Prompt] 正在生成负向条件 / Generating negative condition...
✅ [Prompt] 提示词条件生成完成 / Prompt generated successfully
🚀 开始采样过程/Starting Sampling...
📊 总步数/Total Steps: 4
Executing node 174, title: Image Resize Sum Restore, class type: Image_Resize_sum_restore
Executing node 239, title: Image Uncrop By Mask (Swwan), class type: ImageUncropByMask
Executing node 193, title: 💾 Save Image Extended 2.88, class type: SaveImageExtended
0%| | 0/4 [00:00<?, ?it/s]
25%|██▌ | 1/4 [00:00<00:01, 2.14it/s]
50%|█████ | 2/4 [00:01<00:01, 1.91it/s]
75%|███████▌ | 3/4 [00:01<00:00, 1.85it/s]
100%|██████████| 4/4 [00:02<00:00, 1.82it/s]
100%|██████████| 4/4 [00:02<00:00, 1.86it/s]
Prompt executed in 3.73 seconds
outputs: {'237': {'text': ['False']}, '193': {'images': []}}
====================================
ComfyUI-0001.webp
Version Details
- Version ID
6148cfddbe79750025d9a77c0a3922270f503ca0f4fe451807c774a99e793347- Version Created
- December 29, 2025