Skip to content

Commit ffe40d2

Browse files
committed
docs: 更新README和示例文档,添加使用示例和许可信息
在README和示例文档中添加了使用示例和许可信息,以便用户更好地理解工具的实际应用场景和许可条款。同时,删除了旧的FAQ文件并将其内容整合到新的FAQ文档中。
1 parent 62f0599 commit ffe40d2

File tree

5 files changed

+116
-10
lines changed

5 files changed

+116
-10
lines changed

README.md

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -6,6 +6,10 @@ https://github.com/user-attachments/assets/95bf9e8d-0c97-442d-8068-a5d27e094f18
66

77
AI-PP3 is an advanced command-line tool that combines multimodal AI analysis with RawTherapee's processing engine to create optimized development profiles (.pp3 files) for RAW photography. Designed for both automated workflows and creative exploration.
88

9+
## Usage Examples
10+
11+
See [example documentation](examples/example.md) for practical usage scenarios.
12+
913
## Table of Contents
1014
- [Key Features](#key-features)
1115
- [Compatibility](#compatibility)

README.zh-CN.md

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -6,6 +6,10 @@ https://github.com/user-attachments/assets/95bf9e8d-0c97-442d-8068-a5d27e094f18
66

77
AI-PP3 是一款结合多模态AI分析与RawTherapee处理引擎的命令行工具,专注于为RAW摄影优化生成PP3配置文件。支持自动化工作流和创意探索。
88

9+
## 使用示例
10+
11+
查看[示例文档](examples/example.zh-CN.md)获取实际使用场景。
12+
913
## 目录
1014

1115
- [主要特性](#主要特性)

examples/example.md

Lines changed: 19 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -5,6 +5,7 @@ This project compares AI-generated RawTherapee processing profiles (PP3) against
55
## Project Structure
66

77
Each example folder contains:
8+
89
- Original RAW file from challenge
910
- Base PP3 (created by opening RAW in RawTherapee without edits)
1011
- `ai.pp3` (AI-generated processing profile)
@@ -23,22 +24,24 @@ Each example folder contains:
2324

2425
## Challenge Images Overview
2526

26-
| Filename | Key Characteristics | Dynamic Range |
27-
|---------------------------|------------------------------------------------------|---------------|
28-
| IMG_0080.CR2 | Sunset, underexposed, high noise, 12Ev | 12 EV |
29-
| 2010_MONTR_033.NEF | Mixed lighting ("Blue Horse"), deep whites/blacks | - |
30-
| S7_00463.ARW | Low-light dance scene, artificial lighting, 16Ev | 16 EV |
31-
| B.dng | Dull countryside scene, low contrast | - |
32-
| IMGP2426.DNG | Sunset preservation, extreme underexposure, 14Ev | 14 EV |
33-
| DSC5286.NEF | Cruise ship detail, motion blur, low contrast | - |
27+
| Filename | Key Characteristics | Dynamic Range |
28+
| ------------------ | ------------------------------------------------- | ------------- |
29+
| IMG_0080.CR2 | Sunset, underexposed, high noise, 12Ev | 12 EV |
30+
| 2010_MONTR_033.NEF | Mixed lighting ("Blue Horse"), deep whites/blacks | - |
31+
| S7_00463.ARW | Low-light dance scene, artificial lighting, 16Ev | 16 EV |
32+
| B.dng | Dull countryside scene, low contrast | - |
33+
| IMGP2426.DNG | Sunset preservation, extreme underexposure, 14Ev | 14 EV |
34+
| DSC5286.NEF | Cruise ship detail, motion blur, low contrast | - |
3435

3536
## AI Processing Workflow
3637

3738
1. **Base PP3 Creation**:
39+
3840
- Open RAW file in RawTherapee 5.9+
3941
- Immediately save as base.pp3 without adjustments
4042

4143
2. **AI Profile Generation**:
44+
4245
```bash
4346
# Example from /examples/1/ai.sh
4447
#!/bin/bash
@@ -52,13 +55,15 @@ ai-pp3 IMG_0080.CR2 \
5255
```
5356

5457
**Requirements**:
58+
5559
- RawTherapee 5.9+
5660
- OpenRouter API key (for Gemini access)
5761
- Basic command-line knowledge
5862

5963
## Evaluation Methodology
6064

6165
1. **Technical Comparison**:
66+
6267
```bash
6368
# Structural diff ignoring timestamps
6469
diff -I '\<Date\>' base.pp3 ai.pp3
@@ -72,8 +77,12 @@ awk -F= '/^[^#]/ {print $1}' ai.pp3 | sort | uniq -c
7277
- Histogram analysis of tone distribution
7378
- Zoomed inspection of noise patterns
7479

80+
## License Information
81+
82+
All image materials in this folder are licensed under the CC BY-NC-SA 4.0 license.
83+
7584
## Reference Materials
7685

77-
- [RawTherapee Processing Challenge](https://rawpedia.rawtherapee.com/Rawtherapee_Processing_Challenge_feedback)
78-
- [PP3 Sidecar Documentation](https://rawpedia.rawtherapee.com/Sidecar_Files_-_Processing_Profiles)
86+
- [RawTherapee Processing Challenge](https://rawpedia.rawtherapee.com/Rawtherapee_Processing_Challenge_feedback)
87+
- [PP3 Sidecar Documentation](https://rawpedia.rawtherapee.com/Sidecar_Files_-_Processing_Profiles)
7988
- [OpenRouter API](https://openrouter.ai/docs)

examples/example.zh-CN.md

Lines changed: 89 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,89 @@
1+
# AI-PP3 处理挑战分析
2+
3+
本项目将基于 Gemini 2.5 Pro 生成的 AI 自动化 RawTherapee 处理配置文件(PP3)与 [RawTherapee 处理挑战](https://rawpedia.rawtherapee.com/Rawtherapee_Processing_Challenge_feedback) 中的人工编辑结果进行对比。
4+
5+
## 项目结构
6+
7+
每个示例文件夹包含以下内容:
8+
9+
- 挑战提供的原始 RAW 文件
10+
- 基础 PP3 文件(通过直接打开 RAW 文件后未做任何调整即保存生成)
11+
- `ai.pp3`(AI 生成的处理配置文件)
12+
- `ai.jpg`(应用 AI PP3 后的处理结果)
13+
- `ai.sh`(通过 OpenRouter 调用 Gemini 2.5 Pro 生成 PP3 的脚本)
14+
15+
```plaintext
16+
/1
17+
├── IMG_0080.CR2
18+
├── IMG_0080.CR2.pp3
19+
├── ai.pp3
20+
├── ai.jpg
21+
└── ai.sh
22+
... (共 6 个示例文件夹)
23+
```
24+
25+
## 挑战图像概览
26+
27+
| 文件名 | 核心特征 | 动态范围 |
28+
| ------------------ | ------------------------------------- | -------- |
29+
| IMG_0080.CR2 | 日落场景、欠曝、高噪点、12EV 动态范围 | 12 EV |
30+
| 2010_MONTR_033.NEF | 混合光线("蓝马")、深白/黑区域 | - |
31+
| S7_00463.ARW | 低光舞会场景、人工照明、16EV 动态范围 | 16 EV |
32+
| B.dng | 阴郁乡村场景、低对比度 | - |
33+
| IMGP2426.DNG | 日落场景保护、极端欠曝、14EV 动态范围 | 14 EV |
34+
| DSC5286.NEF | 轮船细节、运动模糊、低对比度 | - |
35+
36+
## AI 处理流程
37+
38+
### 1. 基础 PP3 创建
39+
40+
- 在 RawTherapee 5.9+ 中打开 RAW 文件
41+
- 无需任何调整直接保存为 `base.pp3`
42+
43+
### 2. AI 配置文件生成
44+
45+
```bash
46+
# 示例:/examples/1/ai.sh
47+
#!/bin/bash
48+
export OPENAI_API_KEY=your-key-here
49+
50+
ai-pp3 IMG_0080.CR2 \
51+
--provider openrouter \
52+
--model google/gemini-pro-1.5 \
53+
--base IMG_0080.CR2.pp3 \
54+
--output ai.pp3
55+
```
56+
57+
**依赖项**
58+
59+
- RawTherapee 5.9+
60+
- OpenRouter API 密钥(用于调用 Gemini 模型)
61+
- 基础命令行操作能力
62+
63+
## 评估方法
64+
65+
### 1. 技术对比
66+
67+
```bash
68+
# 忽略时间戳的结构化差异
69+
diff -I '\<Date\>' base.pp3 ai.pp3
70+
71+
# 参数统计
72+
awk -F= '/^[^#]/ {print $1}' ai.pp3 | sort | uniq -c
73+
```
74+
75+
### 2. 视觉评估
76+
77+
- RawTherapee 中并排对比
78+
- 直方图分析色调分布
79+
- 放大观察噪点模式
80+
81+
## 许可声明
82+
83+
本文件夹中的所有图像素材均采用知识共享署名-非商业性使用-相同方式共享 4.0 国际许可协议(CC BY-NC-SA 4.0)授权。
84+
85+
## 参考资料
86+
87+
- [RawTherapee 处理挑战](https://rawpedia.rawtherapee.com/Rawtherapee_Processing_Challenge_feedback)
88+
- [PP3 辅助配置文件文档](https://rawpedia.rawtherapee.com/Sidecar_Files_-_Processing_Profiles)
89+
- [OpenRouter API 文档](https://openrouter.ai/docs)
File renamed without changes.

0 commit comments

Comments
 (0)