Patch-Based Spatial Authorship Attribution in Human-Robot Collaborative Paintings
This addresses the need for documenting authorship in creative AI workflows for artists, collectors, and legal contexts, though it is specific to a particular human-robot pair and thus incremental in scope.
The paper tackles the problem of spatial authorship attribution in human-robot collaborative paintings by developing a patch-based framework, achieving 88.8% patch-level accuracy and using conditional Shannon entropy to quantify stylistic overlap with 64% higher uncertainty in hybrid regions.
As agentic AI becomes increasingly involved in creative production, documenting authorship has become critical for artists, collectors, and legal contexts. We present a patch-based framework for spatial authorship attribution within human-robot collaborative painting practice, demonstrated through a forensic case study of one human artist and one robotic system across 15 abstract paintings. Using commodity flatbed scanners and leave-one-painting-out cross-validation, the approach achieves 88.8% patch-level accuracy (86.7% painting-level via majority vote), outperforming texture-based and pretrained-feature baselines (68.0%-84.7%). For collaborative artworks, where ground truth is inherently ambiguous, we use conditional Shannon entropy to quantify stylistic overlap; manually annotated hybrid regions exhibit 64% higher uncertainty than pure paintings (p=0.003), suggesting the model detects mixed authorship rather than classification failure. The trained model is specific to this human-robot pair but provides a methodological grounding for sample-efficient attribution in data-scarce human-AI creative workflows that, in the future, has the potential to extend authorship attribution to any human-robot collaborative painting.