ScribbleEdit: A Benchmark for Scribble-Only Image Editing

Key points

• To systematically study this problem, we construct a new benchmark, ScribbleEdit, that evaluates the ability of image editing models to perform image editing conditioned on scribbles.1
• In ScribbleEdit, we design an automated data construction pipeline and introduce a dedicated evaluation protocol that explicitly measures intention alignment.1
• Our analysis reveals that existing VLM/LLM-based editing models fail to accurately capture scribble intentions.1
• To guide future progress on scribble-only image editing, we propose a simple yet effective soft-token baseline, which enhances the model's understanding of scribble semantics and outperforms standard image editing models on our benchmark.1

Sources (1)

• [1]ScribbleEdit: A Benchmark for Scribble-Only Image Editing
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 03:38 AM

  Scribble-based interaction provides a lightweight and intuitive way for users to specify image editing intents in interactive editing tools.

  To systematically study this problem, we construct a new benchmark, ScribbleEdit, that evaluates the ability of image editing models to perform image editing conditioned on scribbles.

Extractive summary: sentences quoted from the sources.