SVGauge: Towards Human-Aligned Evaluation for SVG Generation
This addresses the need for vector-specific evaluation in SVG generation, providing a practical tool for benchmarking future models, though it is incremental as it builds on existing embedding and captioning methods.
The paper tackles the problem of evaluating text-to-SVG generation by introducing SVGauge, a human-aligned metric that measures visual fidelity and semantic consistency, achieving the highest correlation with human judgments and faithfully reproducing system-level rankings on the SHE benchmark.
Generated Scalable Vector Graphics (SVG) images demand evaluation criteria tuned to their symbolic and vectorial nature: criteria that existing metrics such as FID, LPIPS, or CLIPScore fail to satisfy. In this paper, we introduce SVGauge, the first human-aligned, reference based metric for text-to-SVG generation. SVGauge jointly measures (i) visual fidelity, obtained by extracting SigLIP image embeddings and refining them with PCA and whitening for domain alignment, and (ii) semantic consistency, captured by comparing BLIP-2-generated captions of the SVGs against the original prompts in the combined space of SBERT and TF-IDF. Evaluation on the proposed SHE benchmark shows that SVGauge attains the highest correlation with human judgments and reproduces system-level rankings of eight zero-shot LLM-based generators more faithfully than existing metrics. Our results highlight the necessity of vector-specific evaluation and provide a practical tool for benchmarking future text-to-SVG generation models.