• We are thrilled to announce that our paper has been accepted for IROS24 in Abu Dhabi!
• We will present our paper as both an oral presentation and a poster session at IROS24.

The integration of robotics and artificial intelligence has paved the way for innovative solutions in various domains. This paper introduces PICaSo (Physical Inpainting on Canvas Solution), a pioneering robotic painting system driven by fine-tuned text-to-image models. Departing from conventional approaches, PICaSo harnesses the power of tailored text-to-image algorithms to interpret natural language prompts and execute precise artistic renderings on canvas. Users guide the process by furnishing descriptive text, specifying desired imagery and placement, empowering the robotic arm to autonomously translate these instructions into physical artworks. Our system's innovation lies in its effective translation of digital inpainting processes into physical actions. Leveraging our multi-tools gripper, capable of seamlessly switching between eraser and pen functions, enables to seamlessly execute detailed drawings on canvas while also providing the flexibility to erase and redo sections as required. This paper comprehensively outlines the capabilities of the proposed system, explores potential applications across various domains, and addresses technical challenges encountered during its development.