ICML2025

MindCustomer: Multi-Context Image Generation Blended with Brain Signal

Muzhou Yu, Shuyun Lin, Lei Ma, Bo Lei, Kaisheng Ma

摘要

Figure 1. MindCustomer seamlessly integrates visual brain signals into multi-context image generation. Given the scarcity of real-world brain data, it efficiently performs real-time single-image generation without additional inputs (e.g. masks or muti-views of the same image context). Text context is optional, which provides more semantics or control over the image and brain contexts.