[CVPR 2025] Official repository of the paper "Mask-Adapter: The Devil is in the Masks for Open-Vocabulary Segmentation"
active 2024-12-08 → 2026-03-10 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 458 days from 2024-12-08 to 2026-03-10. Pushes: 9 total, peak 2 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 34 total, peak 3 in a day. Comments: 49 total, peak 12 in a day. Stars: 104 total, peak 11 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 |
|---|---|---|---|---|
| owl-10 | 42 | 5 | 0 | 28 |
| zzzketi | 8 | 0 | 0 | 7 |
| ShunZuo-AI | 6 | 0 | 0 | 4 |
| DailyVy | 5 | 0 | 0 | 1 |
| wondervictor | 4 | 4 | 0 | 0 |
| colorfulandcjy0806 | 4 | 0 | 0 | 2 |
| UyJZ | 3 | 0 | 0 | 1 |
| snowpine-max | 3 | 0 | 0 | 1 |
| so-ux | 2 | 0 | 0 | 1 |
| freeLGR | 2 | 0 | 0 | 1 |
| Jiang-Shiqi123 | 2 | 0 | 0 | 0 |
| Michelvl92 | 2 | 0 | 0 | 1 |
| ArsenLuca | 2 | 0 | 0 | 1 |
| NielsRogge | 2 | 0 | 0 | 1 |
| LMerCy | 1 | 0 | 0 | 0 |
| c1ircle | 1 | 0 | 0 | 0 |
| MUVGuan | 1 | 0 | 0 | 0 |
| sajjad-sh33 | 1 | 0 | 0 | 0 |
| DanilaAniva | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#20owl-102025-12-31 06:38Evaluation performance of released Mask-Adapter ConvNeXt-L checkpoints lower than reported
- Issue#21Jiang-Shiqi1232025-12-29 03:27First-stage training epochs requirement for MAFTP-Large before second-stage
- Issue#21Jiang-Shiqi1232025-12-29 02:27First-stage training epochs requirement for MAFTP-Large before second-stage
- Issue#19owl-102025-12-25 17:16Inference Time
- Issue comment#19owl-102025-12-25 17:16Inference Time
- Issue comment#20owl-102025-12-25 17:01Evaluation performance of released Mask-Adapter ConvNeXt-L checkpoints lower than reported
- Issue#7owl-102025-10-05 16:27where is the demo script?
- Issue comment#18owl-102025-08-11 10:02FC-CLIP + Mask-Adapter 第二阶段训练后无法达到论文里的数值
- Issue comment#18owl-102025-07-08 02:31FC-CLIP + Mask-Adapter 第二阶段训练后无法达到论文里的数值
- Issue comment#18owl-102025-07-04 13:37FC-CLIP + Mask-Adapter 第二阶段训练后无法达到论文里的数值
- Issue#14owl-102025-06-30 02:47Question Regarding Ground-Truth Mask Usage During Training and Potential Information Leakage
- Issue#16owl-102025-05-28 08:05semantic activation maps generation
- Issue comment#9Michelvl922025-05-22 11:54Unable to Run MaskAdapter with MAFT+
- Issue comment#16so-ux2025-05-20 06:09semantic activation maps generation
- Issue comment#16owl-102025-05-20 02:47semantic activation maps generation
- Issue#12owl-102025-05-19 14:01关于demo.py使用fcclip_convnext_large_maskadapter.pth进行图片推理的问题
- Issue comment#17zzzketi2025-05-19 13:41Significant Gap Between Reproduced Results and Reported Performance in Paper
- Issue comment#17owl-102025-05-19 13:12Significant Gap Between Reproduced Results and Reported Performance in Paper
- Issue comment#17zzzketi2025-05-19 13:11Significant Gap Between Reproduced Results and Reported Performance in Paper
- Issue comment#17owl-102025-05-19 13:09Significant Gap Between Reproduced Results and Reported Performance in Paper
- Issue comment#17zzzketi2025-05-19 13:02Significant Gap Between Reproduced Results and Reported Performance in Paper
- Issue comment#17zzzketi2025-05-19 12:56Significant Gap Between Reproduced Results and Reported Performance in Paper
- Issue comment#17zzzketi2025-05-19 12:44Significant Gap Between Reproduced Results and Reported Performance in Paper
- Issue comment#17owl-102025-05-19 12:11Significant Gap Between Reproduced Results and Reported Performance in Paper
- Issue comment#17zzzketi2025-05-19 12:00Significant Gap Between Reproduced Results and Reported Performance in Paper
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 104 stars here means stars gained during the window, not the repo's star count.