Update README.md
Browse files
README.md
CHANGED
@@ -15,8 +15,8 @@ A sophisticated AI-powered image generation system that creates high-quality mar
|
|
15 |
|
16 |
## Features
|
17 |
|
18 |
-
- **AI-Powered Image Generation**: Create stunning marketing images from text prompts using Google's
|
19 |
-
- **Automated Quality Review**: Intelligent Gemini agent automatically reviews and refines generated images
|
20 |
- **Marketing-Focused**: Optimized for marketing materials, social media, and promotional content
|
21 |
- **Real-time Feedback**: Get instant quality scores and improvement suggestions
|
22 |
- **Professional Workflow**: Streamlined process from concept to final image
|
@@ -74,7 +74,7 @@ A sophisticated AI-powered image generation system that creates high-quality mar
|
|
74 |
βPrompt β β β β Reviewer β
|
75 |
β β β β β β
|
76 |
β β β β β ββββββββββββββββββββββββββββ
|
77 |
-
β β β β β β Ag1:
|
78 |
β β β β β β ββ
|
79 |
β β β β β β Draft Image Creation ββ
|
80 |
β β β β β ββββββββββββββββββββββββββββ
|
@@ -104,7 +104,7 @@ A sophisticated AI-powered image generation system that creates high-quality mar
|
|
104 |
|
105 |
3. **Image Generation and Drafting (Top Right)**:
|
106 |
- **Agent 1 (Gemini) Drafter**: Receives Image Prompt, orchestrates image generation
|
107 |
-
- **Imagen3 (via MCP)**: Agent 1 interacts with
|
108 |
|
109 |
4. **Marketing Review and Refinement (Bottom Right)**:
|
110 |
- **Agent 2 (Gemini) Marketing Reviewer**: Receives Reviewer Prompt, evaluates generated image against marketing criteria
|
@@ -113,11 +113,11 @@ A sophisticated AI-powered image generation system that creates high-quality mar
|
|
113 |
- Final **Image Response** sent back to Gradio UI
|
114 |
|
115 |
### Summary of Flow:
|
116 |
-
User provides prompts β Gradio UI β Agent 1 drafts image with
|
117 |
|
118 |
### Technology Stack
|
119 |
|
120 |
-
- **AI Models**: Google
|
121 |
- **Framework**: Gradio (Web Interface)
|
122 |
- **Orchestration**: Custom agent handover system
|
123 |
- **Deployment**: Hugging Face Spaces
|
@@ -129,7 +129,7 @@ User provides prompts β Gradio UI β Agent 1 drafts image with Imagen3 β Ag
|
|
129 |
The system was designed with a **custom handover mechanism** instead of the A2A (Agent-to-Agent) protocol for the following reasons:
|
130 |
|
131 |
1. **Simplified Architecture**: The current two-agent system (generator + reviewer) doesn't require the complexity of full A2A orchestration
|
132 |
-
2. **Direct Integration**: MCP server provides direct access to
|
133 |
3. **Performance Optimization**: Direct handover between agents reduces latency and eliminates protocol overhead
|
134 |
4. **Deployment Simplicity**: Hugging Face Spaces deployment is more straightforward without A2A dependencies
|
135 |
5. **Resource Efficiency**: Fewer moving parts means better resource utilization in the cloud environment
|
@@ -287,7 +287,7 @@ This project is licensed under the MIT License - see the LICENSE file for detail
|
|
287 |
|
288 |
## Acknowledgments
|
289 |
|
290 |
-
- Google AI for
|
291 |
- Hugging Face for the deployment platform
|
292 |
- Gradio for the web interface framework
|
293 |
- The open-source community for various dependencies
|
|
|
15 |
|
16 |
## Features
|
17 |
|
18 |
+
- **AI-Powered Image Generation**: Create stunning marketing images from text prompts using Google's Imagen4 via MCP server
|
19 |
+
- **Automated Quality Review**: Intelligent Gemini agent (2.5-Pro) automatically reviews and refines generated images
|
20 |
- **Marketing-Focused**: Optimized for marketing materials, social media, and promotional content
|
21 |
- **Real-time Feedback**: Get instant quality scores and improvement suggestions
|
22 |
- **Professional Workflow**: Streamlined process from concept to final image
|
|
|
74 |
βPrompt β β β β Reviewer β
|
75 |
β β β β β β
|
76 |
β β β β β ββββββββββββββββββββββββββββ
|
77 |
+
β β β β β β Ag1: Imagen4 (via MCP) ββ
|
78 |
β β β β β β ββ
|
79 |
β β β β β β Draft Image Creation ββ
|
80 |
β β β β β ββββββββββββββββββββββββββββ
|
|
|
104 |
|
105 |
3. **Image Generation and Drafting (Top Right)**:
|
106 |
- **Agent 1 (Gemini) Drafter**: Receives Image Prompt, orchestrates image generation
|
107 |
+
- **Imagen3 (via MCP)**: Agent 1 interacts with Imagen4 through MCP server to create initial image draft
|
108 |
|
109 |
4. **Marketing Review and Refinement (Bottom Right)**:
|
110 |
- **Agent 2 (Gemini) Marketing Reviewer**: Receives Reviewer Prompt, evaluates generated image against marketing criteria
|
|
|
113 |
- Final **Image Response** sent back to Gradio UI
|
114 |
|
115 |
### Summary of Flow:
|
116 |
+
User provides prompts β Gradio UI β Agent 1 drafts image with Imagen4 β Agent 2 reviews and suggests refinements β Iterative refinement loop β Final reviewed image β User receives result
|
117 |
|
118 |
### Technology Stack
|
119 |
|
120 |
+
- **AI Models**: Google Imagen4 (via MCP), Gemini Vision
|
121 |
- **Framework**: Gradio (Web Interface)
|
122 |
- **Orchestration**: Custom agent handover system
|
123 |
- **Deployment**: Hugging Face Spaces
|
|
|
129 |
The system was designed with a **custom handover mechanism** instead of the A2A (Agent-to-Agent) protocol for the following reasons:
|
130 |
|
131 |
1. **Simplified Architecture**: The current two-agent system (generator + reviewer) doesn't require the complexity of full A2A orchestration
|
132 |
+
2. **Direct Integration**: MCP server provides direct access to Imagen4 without needing agent-to-agent communication protocols
|
133 |
3. **Performance Optimization**: Direct handover between agents reduces latency and eliminates protocol overhead
|
134 |
4. **Deployment Simplicity**: Hugging Face Spaces deployment is more straightforward without A2A dependencies
|
135 |
5. **Resource Efficiency**: Fewer moving parts means better resource utilization in the cloud environment
|
|
|
287 |
|
288 |
## Acknowledgments
|
289 |
|
290 |
+
- Google AI for Imagen4 and Gemini technologies
|
291 |
- Hugging Face for the deployment platform
|
292 |
- Gradio for the web interface framework
|
293 |
- The open-source community for various dependencies
|