[ECCV 2024] The official implementation of paper "BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion"
Python · active 2024-03-10 → 2026-04-18 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 770 days from 2024-03-10 to 2026-04-18. Pushes: 16 total, peak 3 in a day. Pull requests: 9 total, peak 2 in a day. Issues: 109 total, peak 7 in a day. Comments: 176 total, peak 9 in a day. Stars: 1,688 total, peak 119 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| juxuan27 | 83 | 16 | 4 | 43 |
| Shuvo001 | 15 | 0 | 0 | 12 |
| xduzhangjiayu | 15 | 0 | 2 | 12 |
| cs-mshah | 13 | 0 | 0 | 8 |
| CharlesGong12 | 8 | 0 | 0 | 6 |
| henbucuoshanghai | 7 | 0 | 0 | 4 |
| Arcxml | 6 | 0 | 0 | 5 |
| lawsonxwl | 6 | 0 | 0 | 5 |
| dydxdt | 4 | 0 | 0 | 2 |
| MasaiahHan | 4 | 0 | 0 | 4 |
| PlutoQyl | 4 | 0 | 0 | 1 |
| SlZeroth | 3 | 0 | 0 | 2 |
| MrWH123 | 3 | 0 | 0 | 3 |
| ypflll | 3 | 0 | 0 | 2 |
| MikoSamey | 3 | 0 | 0 | 3 |
| zhanjiahui | 3 | 0 | 0 | 2 |
| huangjun12 | 3 | 0 | 0 | 0 |
| RunpuWei | 3 | 0 | 0 | 3 |
| Chevolier | 3 | 0 | 0 | 1 |
| huiyang865 | 3 | 0 | 0 | 3 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#85Marcus20652025-09-16 11:18Error message occurred while importing the 'ComfyUI-BrushNet' module.
- Issue comment#84Marcus20652025-09-16 11:18安装了brushnet最新版启动报错
- Issue comment#51Kuanshen772025-06-30 07:25About the Mask and Masked Image pair. Difference between training and test codes.
- Issue comment#78akhilgvss2025-06-27 22:20Unable to reproduce the paper results
- Issue comment#29huangjie-nlp2025-06-23 03:31Create Custom BrushData set.
- Issue comment#29huangjie-nlp2025-06-23 03:30Create Custom BrushData set.
- Issue comment#39santisy2025-05-20 03:40Errors when training with dataloader_num_workers > 0
- Issue comment#35santisy2025-05-17 23:42Finetune SDXL
- Issue comment#83alexrabbit972025-05-10 15:59测试的时候导包错误
- Issue#86zhanjiahui2025-05-07 14:43A note about BrushData!
- Issue#85gg1012012025-04-27 18:08Error message occurred while importing the 'ComfyUI-BrushNet' module.
- Issue comment#76guajioao2025-04-23 00:42增加额外的layer by layer特征，逐层和原始unet求和。那这些额外的特征参数量和unet一一对应？同样是个大模型？
- Issue#84kileroppo2025-04-21 23:00安装了brushnet最新版启动报错
- Issue#83HarlayZhang2025-03-25 14:01测试的时候导包错误
- Issue comment#75zhanjiahui2025-03-24 11:14蒙版不能是方形的嘛？
- Issue#82huangjun122025-03-22 03:35The LPIPS and MSE metrics of BrushBench seem to be reversed in the paper？
- Issue#82huangjun122025-03-21 16:23The LPIPS and MSE metrics of BrushBench and EditBench seem to be reversed in the paper？
- Issue comment#34zhanjiahui2025-03-07 14:34why brushnet control need noise_latent as part of condition ?
- Issue comment#44qiqiApink2025-03-06 11:22is the traindata model input correct？
- Issue#79htyjers2025-02-22 03:31Quantitative result
- Issue#81chaledelta2025-02-18 05:37CUDA out of memory.How much memory is needed for training?（训练时需要多大的显存？）
- Issue comment#79httang12242025-02-04 14:18Quantitative result
- Issue comment#74josiahZXY2025-01-14 12:23EditBench not found in google drive
- Issue#80juxingyiwan2025-01-10 14:04Training erroe with accelerate.backward(loss)
- Issue#79htyjers2025-01-09 03:10Quantitative result
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 1,688 stars here means stars gained during the window, not the repo's star count.