Release v3.0.0

What's Changed

• Structured response in #35

Full Changelog: v2.0.0...v3.0.0

Release v2.0.0

Changelog

• GUI Implementation: Intuitive graphical user interface for easy interaction with the platform.
• Single Analysis Mode: Analyze individual images with customizable distortion effects and prompts.
• Bulk Analysis Mode: Process multiple images simultaneously with customizable settings.
• Centralized Distortion Settings: Apply uniform distortion effects across all images in bulk analysis.
• Folder Path Input: Easily analyze multiple images by specifying a folder path.
  - Enhanced Image Distortions:
  • Color Adjustment: Fine-tune image saturation and hue.
  • Overlay Effect: Apply custom overlay images with adjustable intensity.
  • Warp Effect: Create wave and bulge distortions with customizable parameters.
• Customizable System Instructions: Tailor the AI's behavior with user-defined instructions.
• Progress Tracking: Monitor the progress of bulk analysis operations.
• CSV Export: Download analysis results in a structured CSV format for further processing.

This release marks the first GUI implementation of our Multimodal LLM Road Safety Platform, offering both single and bulk analysis modes. It significantly expands the capabilities of the platform, providing more flexibility and power in analyzing road safety scenarios at various scales.

Release v1.0.0

Changelog

• Implemented the CLI interface.
• Integrated AI functionality into the CLI using the Gemini 1.5 Flash model for real-time processing.