Vibe Spaces for Creatively Connecting and Expressing Visual Concepts
This addresses a challenge in creative visual generation for artists and designers, though it appears incremental as it builds on existing feature spaces like CLIP.
The paper tackles the problem of generating coherent visual concept hybrids by identifying shared attributes between images, introducing Vibe Space to enable smooth transitions in latent spaces like CLIP, and finds that it produces blends rated as more creative and coherent by humans compared to current methods.
Creating new visual concepts often requires connecting distinct ideas through their most relevant shared attributes -- their vibe. We introduce Vibe Blending, a novel task for generating coherent and meaningful hybrids that reveals these shared attributes between images. Achieving such blends is challenging for current methods, which struggle to identify and traverse nonlinear paths linking distant concepts in latent space. We propose Vibe Space, a hierarchical graph manifold that learns low-dimensional geodesics in feature spaces like CLIP, enabling smooth and semantically consistent transitions between concepts. To evaluate creative quality, we design a cognitively inspired framework combining human judgments, LLM reasoning, and a geometric path-based difficulty score. We find that Vibe Space produces blends that humans consistently rate as more creative and coherent than current methods.