Get Started
Home
Topics
Search
Library
Research questionHow can generative image systems preserve subject identity under viewpoint changes, degradation, and iterative edits?A generative system can produce visually compelling images and follow instructions while changing the identity of the depicted subject. This drift becomes more pronounced with viewpoint changes, small or degraded subjects, repeated edits, and multiple subjects.
AI
Computer Vision
Evaluation & Benchmarks
Image & Video Processing
Image Generation
Machine Learning
Research Paper
Latest papersRecent research connected to this question, newest first.Persistent Identity Preservation in Generative Image Models: A Benchmark and Evaluation SystemThe evidence covers benchmarks comparing input-context representations, trainable subject-specific parameters, and persistent identity layers across generation, editing, restoration, and multi-subject tasks. It reports identity preservation alongside instruction adherence and perceptual image quality across different foundation models.research paper · Sep 10, 2026CanvasComposer: Personalized Group Photo Generation via a Multi-Reference CanvasThe source concerns an interactive framework where users place separate RGBA subject cutouts on a shared canvas before generating a single harmonized group image. Reported evidence addresses identity preservation and visual coherence in multi-human personalized image generation.research paper · Sep 2, 2026
Related questions
How can image editors infer edit regions and preserve non-target content without introducing artifacts?How can open image generators follow fine-grained editing instructions without sacrificing image quality or inference speed?How can composed image retrieval distinguish changed, preserved, and removed visual details?How can single-image diffusion personalization preserve identity while enabling controllable attribute-level context changes?