cs.CVNov 6, 2026

Same Pixels, Different Documents: Measuring Part-Level Editability in Image Vectorization

Authors: Boheng Zhu, Zehua Wang, Wei Chen, Ming-Hao Lin

Organizations: China University of Mining and Technology · The University of British Columbia · Concealed Studio

Abstract

Image vectorization is often evaluated by rendering fidelity, which cannot distinguish documents that render identically but support different edits. Edit-based evaluation tests outcomes, but an oracle that gathers a part's fragments can conceal the cost of selecting them. We introduce Pass@k, which tests whether a prescribed edit can succeed using objects selectable within k clicks under a specified interaction model. Eight document diagnostics characterize properties relevant to editing beyond these tasks. We assess structural sensitivity through controlled perturbations, practical relevance through designer operations, and reference dependence through alternative drawings. We evaluate twelve open and commercial methods on 211 references, with 635 edit tasks on 93 of them. In each of four reference sets, including two drawn by designers, 64-88% of oracle-passed tasks on parts composed of multiple shapes fail at Pass@1, against 6-40% for single shapes, pooling counts over configurations. Higher fidelity does not guarantee higher Pass@1, while selected diagnostics are associated with designers' effort in our study.

Figures & tables

Explore similar work

CardsList
  1. Vector-Bench: Can Models Surgically Edit SVG Code?

    Jul 21, 2026Yug Aditi Gupta, Prannay HebbarImage Editing EvaluationImage Editing

  2. Beyond Plausibility: Verifiable Fine-Grained Image Editing on Structured Assets

    Oct 3, 2026Muyao Wang, Chen Zhu, Shiqi Yang +2Image EditingImage Editing Evaluation

  3. VDE Bench: Evaluating The Capability of Image Editing Models to Modify Visual Documents

    Jan 27, 2026Hongzhu Yi, Yujia Yang, Yuanxiang Wang +18Image Editing EvaluationImage Editing