Prompt Description

This image illustrates an automated pipeline for reverse engineering chart images into TikZ code. The core logic of this system can be divided into the following four main stages: 1. Input & Global Planning Stage: Input: Original chart images (such as line charts, bar charts, etc.). Global Processing: Global Planner: Receives image input and extracts global metadata. Output: Generates JSON formatted data containing the chart's global parameters (e.g., 'X-scale': 1.5, 'Y-scale': 0.5, etc.). 2. Parallel Feature Extraction Stage: The image is simultaneously input into four parallel deep learning/recognition models, each responsible for extracting different chart component information and generating corresponding TikZ code snippets: Chart Basic Configuration Recognition Model: Identifies the basic structure of the chart and generates Tikz 1 code. Axis Parsing Model: Parses the scales, labels, etc., of the X/Y axes and generates Tikz 2 code. Data Visualization Recognition Model: Identifies specific data points, lines, or bars and generates Tikz 3 code. Legend Recognition Model: Identifies the content and style of the legend and generates Tikz 4 code. 3. Code Synthesis Stage: Fill-in: Integrates the data produced in the above two stages: JSON data produced by the Global Planner. TikZ code snippets (1-4) produced by the four models. Template Application: This data is filled into a preset LaTeX template. Output: Generates complete TikZ code. 4. Compilation & Validation Loop: This is a crucial closed-loop in the system to ensure code executability: Latex Compiler: Attempts to compile the generated complete code. Success: Directly outputs the final chart, and the process ends. Failure: Triggers an error handling mechanism. Check Agent: Receives compilation failure information. Analyzes the error and attempts to fix the code. Generates New TikZ Code. Loop: Sends the new code back to the Latex Compiler for compilation until successful.