ovi054 commited on
Commit
5f338cb
Β·
verified Β·
1 Parent(s): 0d3cb75

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +17 -2
README.md CHANGED
@@ -33,6 +33,17 @@ short_description: Apply the color grade from one image to another image
33
 
34
  It transfers the color grading from a reference image directly onto a source image.
35
 
 
 
 
 
 
 
 
 
 
 
 
36
  ### Key Features
37
  * **Direct Color Transfer:** Copies styling from a reference image to a source image without manual grading.
38
  * **4-Step Inference:** Uses a lightning adapter for fast, low-step generation.
@@ -43,13 +54,13 @@ It transfers the color grading from a reference image directly onto a source ima
43
  ---
44
 
45
 
46
- ### πŸ› οΈ Core Constraints
47
  * **Small Models Only:** Powered by the `Qwen-Image-Edit-2511` architecture family running at 20B parameters safely satisfying the 32 Billion parameter limit.
48
  * πŸ”Œ **Off the Grid (Local-First):** **No cloud APIs are used.** The pipeline loads the model weights and runs entirely on the host GPU hardware.
49
  * 🎯 **Well-Tuned:** Utilizes a custom fine-tuned LoRA adapter published directly on my Hugging Face Hub (`ovi054/QIE-2511-Color-Grade-Transfer-LoRA`).
50
 
51
  ---
52
-
53
 
54
  ## 🧠 LoRA Fine-Tuning Process
55
 
@@ -102,6 +113,10 @@ The model was fine-tuned using the following precise configuration settings:
102
  ### Live Demo
103
  * **Hugging Face Space:** [Color Grade Transfer](https://huggingface.co/spaces/build-small-hackathon/Color-Grade-Transfer)
104
 
 
 
 
 
105
  ### Social Media & Demo
106
  * **Twitter/X:** [https://x.com/ovi054/status/2064060651706925403](https://x.com/ovi054/status/2064060651706925403)
107
  * **Hugging Face Post:** [https://huggingface.co/posts/ovi054/229194823175599](https://huggingface.co/posts/ovi054/229194823175599)
 
33
 
34
  It transfers the color grading from a reference image directly onto a source image.
35
 
36
+
37
+ ## πŸ† Hackathon Patches
38
+
39
+ | Patch | Status |
40
+ | :--- | :--- |
41
+ | πŸ”¬ **Small Models Only** β€” `Qwen-Image-Edit-2511` at 20B parameters, well under the 32B limit | βœ… |
42
+ | πŸ”Œ **Off the Grid** β€” No cloud APIs, runs entirely on local GPU | βœ… |
43
+ | 🎯 **Well-Tuned** β€” Custom LoRA [`ovi054/QIE-2511-Color-Grade-Transfer-LoRA`](https://huggingface.co/ovi054/QIE-2511-Color-Grade-Transfer-LoRA) published on Hub | βœ… |
44
+ | πŸ““ **Field Notes** β€” [Blog post on what I built and learned](https://huggingface.co/blog/ovi054/color-grade-transfer) | βœ… |
45
+
46
+
47
  ### Key Features
48
  * **Direct Color Transfer:** Copies styling from a reference image to a source image without manual grading.
49
  * **4-Step Inference:** Uses a lightning adapter for fast, low-step generation.
 
54
  ---
55
 
56
 
57
+ <!-- ### πŸ› οΈ Core Constraints
58
  * **Small Models Only:** Powered by the `Qwen-Image-Edit-2511` architecture family running at 20B parameters safely satisfying the 32 Billion parameter limit.
59
  * πŸ”Œ **Off the Grid (Local-First):** **No cloud APIs are used.** The pipeline loads the model weights and runs entirely on the host GPU hardware.
60
  * 🎯 **Well-Tuned:** Utilizes a custom fine-tuned LoRA adapter published directly on my Hugging Face Hub (`ovi054/QIE-2511-Color-Grade-Transfer-LoRA`).
61
 
62
  ---
63
+ -->
64
 
65
  ## 🧠 LoRA Fine-Tuning Process
66
 
 
113
  ### Live Demo
114
  * **Hugging Face Space:** [Color Grade Transfer](https://huggingface.co/spaces/build-small-hackathon/Color-Grade-Transfer)
115
 
116
+ ### Blog
117
+ * **Blog Post:** [Color Grade Transfer β€” Instant Cinematic Tone Matching](https://huggingface.co/blog/ovi054/color-grade-transfer)
118
+
119
+
120
  ### Social Media & Demo
121
  * **Twitter/X:** [https://x.com/ovi054/status/2064060651706925403](https://x.com/ovi054/status/2064060651706925403)
122
  * **Hugging Face Post:** [https://huggingface.co/posts/ovi054/229194823175599](https://huggingface.co/posts/ovi054/229194823175599)