IDraw: Artist Verification from Digital Drawing Images

2026-08-03

AI Summary

This work addresses the challenge of authorship verification in digital drawings when only static images are available—lacking dynamic behavioral data such as pen pressure or stroke dynamics—and when high content similarity across artworks complicates identity discrimination. To this end, we propose a sensor-free verification method that infers latent drawing behaviors from static images and incorporates a content–style disentanglement mechanism to suppress artist-shared content cues. We introduce the first multimodal dataset for digital painting authorship verification and develop a verification framework that fuses an image encoder with handwriting priors. Extensive experiments demonstrate that our approach consistently outperforms standard image-based verification methods across nine mainstream image backbones, reducing verification error rates by up to 40%.

Abstract

As digital drawings are increasingly shared online, reliable authorship verification has become important for protecting artists and resolving disputes. Yet when authorship is questioned, verification may have to rely only on the disputed drawing and reference drawings known to be created by the claimed artist. This setting is challenging for two reasons. First, artist-specific drawing behavior, such as pen pressure and movement speed, is informative but is not available from a completed drawing. Second, similarities in the depicted object or scene can obscure similarities arising from the artist. We propose IDraw, a framework that learns from drawings paired with tablet-pen sensor signals collected from separate training artists. This allows IDraw to infer drawing behavior from completed images during a later authorship dispute, without requiring sensor data from the artist being verified. IDraw also reduces the influence of drawing content by identifying information shared by drawings of the same object across different artists and suppressing it before comparing drawings. To support this approach, we construct the first multimodal dataset for digital drawing authorship verification, containing 1,110 drawings from 37 artists and 14 types of tablet-pen sensor signals. Evaluated on previously unseen artists across nine image-encoder backbones, IDraw consistently outperforms standard image-based verification and reduces verification error by up to 40%. These results demonstrate that inferring drawing behavior from completed images and suppressing drawing content improve digital drawing authorship verification.

Problem

• artist verification
• digital drawing
• authorship verification
• drawing behavior
• content interference

Innovation

• artist verification
• digital drawing
• drawing behavior inference
• content suppression
• multimodal dataset