Get Started
Home
Topics
Search
Library
Research questionHow can a transferable image-style representation separate overall identity from fine-grained visual attributes?Image style combines high-level identity with many visual factors that are entangled with image content. Without an explicit shared representation, styles are difficult to compare, transfer, and model consistently.
Computer Vision
Image & Video Processing
Image Generation
Machine Learning
Latest papersRecent research connected to this question, newest first.MegaStyle++: Scaling Image Style Space through Hierarchical Style DefinitionThe source develops a hierarchical style definition spanning overall style identities and fine-grained visual attributes, and applies it to the MegaStyle++-8M dataset, which contains 150,000 style identities, 1 million fine-grained style prompts, and 8 million stylized images. Its analyses examine the diversity, semantic breadth, and visual-style coverage of this representation.research paper · Sep 2, 2026
Related questions
How can generative image systems preserve subject identity under viewpoint changes, degradation, and iterative edits?How can cross-domain image compositing preserve foreground identity while matching the background’s visual style?How can single-image diffusion personalization preserve identity while enabling controllable attribute-level context changes?How can motion-style transfer learn new styles from few paired examples without artifacts on unseen motions?