CVPR2024: Transfer CLIP for Generalizable Image Denoising
active 2024-05-29 → 2026-01-20 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 602 days from 2024-05-29 to 2026-01-20. Pushes: 1 total, peak 1 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 20 total, peak 2 in a day. Comments: 29 total, peak 4 in a day. Stars: 91 total, peak 5 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 |
|---|---|---|---|---|
| alwaysuu | 23 | 1 | 0 | 16 |
| TwilightArchon | 8 | 0 | 0 | 6 |
| xiaozhuqian | 3 | 0 | 0 | 2 |
| HanzhouLiu | 3 | 0 | 0 | 1 |
| ZhaKiNg | 2 | 0 | 0 | 1 |
| fighterzhao | 2 | 0 | 0 | 1 |
| antonyvan | 2 | 0 | 0 | 1 |
| jiachengzhong | 2 | 0 | 0 | 0 |
| junshaoye | 2 | 0 | 0 | 1 |
| Zzhy2000 | 1 | 0 | 0 | 0 |
| musaeneserkuvan | 1 | 0 | 0 | 0 |
| Feynman1999 | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#12alwaysuu2025-05-27 01:34How do I use this models ?
- Issue comment#12alwaysuu2025-05-25 11:29How do I use this models ?
- Issue#12musaeneserkuvan2025-05-24 20:48How do I use this models ?
- Issue comment#11fighterzhao2025-05-23 02:06image feature similarity based on pre-trained CLIP with ResNet
- Issue comment#11alwaysuu2025-05-23 01:58image feature similarity based on pre-trained CLIP with ResNet
- Issue#11fighterzhao2025-05-23 01:41image feature similarity based on pre-trained CLIP with ResNet
- Issue comment#10ZhaKiNg2025-05-01 08:16how to change LDCT_dataset
- Issue#10ZhaKiNg2025-04-30 13:10how to change LDCT_dataset
- Issue comment#9junshaoye2025-03-24 06:40How can I preprocess low-dose CT images?
- Issue#9alwaysuu2025-03-05 01:33How can I preprocess low-dose CT images?
- Issue comment#9alwaysuu2025-03-01 01:49How can I preprocess low-dose CT images?
- Issue#9junshaoye2025-02-27 13:41How can I preprocess low-dose CT images?
- Issue#8alwaysuu2025-02-10 01:33where can I download the pretrained CLIP model
- Issue comment#8antonyvan2025-02-09 08:14where can I download the pretrained CLIP model
- Issue comment#8alwaysuu2025-02-08 13:29where can I download the pretrained CLIP model
- Issue#8antonyvan2025-02-03 01:05where can I download the pretrained CLIP model
- Issue#7jiachengzhong2025-01-20 06:39Preprocessing of LDCT Images
- Issue comment#7alwaysuu2025-01-19 09:33Preprocessing of LDCT Images
- Issue#6alwaysuu2025-01-19 09:25the rgb range for the clip's encoder, 0~1 or -1~1?
- Issue#7jiachengzhong2025-01-19 09:14Preprocessing of LDCT Images
- Issue comment#6alwaysuu2025-01-17 09:18the rgb range for the clip's encoder, 0~1 or -1~1?
- Issue#6Feynman19992025-01-17 08:59the rgb range for the clip's encoder, 0~1 or -1~1?
- Issue comment#5xiaozhuqian2024-10-18 03:13Hi, the analyses of CLIP image encoder are quite straightforward. First, synthesizing noisy images from clean ones (using Gaussian or poison noise). Then, directly send these noisy and clean images to the CLIP ResNet encoder, without crop, resize or normalization as done in the original CLIP preprocessing. Finally, obtain the dense features of noisy images and clean images from CLIP ResNet encoder, respectively, and compute their similarities (using e.g., cosine distance or CKA similarity).
- Issue comment#5alwaysuu2024-10-17 08:15Hi, the analyses of CLIP image encoder are quite straightforward. First, synthesizing noisy images from clean ones (using Gaussian or poison noise). Then, directly send these noisy and clean images to the CLIP ResNet encoder, without crop, resize or normalization as done in the original CLIP preprocessing. Finally, obtain the dense features of noisy images and clean images from CLIP ResNet encoder, respectively, and compute their similarities (using e.g., cosine distance or CKA similarity).
- Issue comment#5xiaozhuqian2024-09-29 14:43Hi, the analyses of CLIP image encoder are quite straightforward. First, synthesizing noisy images from clean ones (using Gaussian or poison noise). Then, directly send these noisy and clean images to the CLIP ResNet encoder, without crop, resize or normalization as done in the original CLIP preprocessing. Finally, obtain the dense features of noisy images and clean images from CLIP ResNet encoder, respectively, and compute their similarities (using e.g., cosine distance or CKA similarity).
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 91 stars here means stars gained during the window, not the repo's star count.