LatentGandr: Visual Exploration of Generative AI Latent Space via Local Embeddings
This work addresses usability and scalability issues for designers and researchers interacting with generative AI models, representing an incremental improvement over existing visualization methods.
The paper tackles the challenge of navigating high-dimensional latent spaces in generative AI by introducing LatentGandr, a visual analytics technique that uses localized PCA to extract and visualize locally linear dimensions, enabling more efficient exploration and control compared to the state-of-the-art GANSlider.
Generative AI has demonstrated significant potential in creative design, enabling the rapid generation of visual content and imaginative concepts. Although deep AI models achieve effective featurization in the latent space, navigating the space remains a challenge. Current techniques, such as GANSlider and SliderSpace, use multiple sliders to generate high-dimensional vectors in generative AI's latent space. Despite applying (global) PCA to reduce the number of sliders, these approaches struggle with scalability and usability as the number of control dimensions increases. In this paper, we introduce LatentGandr, a visual analytics technique that facilitates latent space exploration by extracting locally linear dimensions from embeddings in high-dimensional latent spaces. By analyzing the topology and local curvature of the embeddings, LatentGandr automatically identifies local neighborhoods and computes their principal components using localized PCA. These local principal components are visualized as interactive image grids, allowing users to efficiently explore and control the generative process, providing an intuitive means to refine the generation of novel content and concepts. To evaluate the effectiveness of LatentGandr, we conducted a study comparing it to GANSlider, the current state-of-the-art visualization interface for generative AI models. The results offer insights into how localized exploration techniques can enhance user interaction with these models.