Subject-Driven Image Generation

Ask about this topic

Overview

These methods condition diffusion models and related generators on reference images so a person, object, or character remains recognizable in new poses, scenes, and styles. Common approaches include fine-tuning with LoRA or textual inversion, reference-image adapters, and training-free attention or feature routing; multi-subject systems also use masks, layouts, and separate identity pathways. Benchmarks increasingly test attribute binding, pose control, and identity retention as scenes grow more complex. A central tension is balancing subject fidelity with prompt alignment and visual diversity, while evaluation remains difficult because global similarity scores can miss identity swaps or collapse.

Latest papers

CardsList

61 papers7 papers in the last four weeks, up 133% on the four weeks before. 0.0% of all new papers.

Oct 23, 2025

CanvasComposer: Personalized Group Photo Generation via a Multi-Reference Canvas

Existing personalized image generators still struggle to preserve multiple reference identities in natural and coherent multi-human generations. To address these limitations, we present CanvasComposer, an interactive framework for personalized group photo generation. Inspired by professional image-editing software, CanvasComposer allows users to place reference subjects on a shared canvas, where each subject keeps its own RGBA cutout of the input. This multi-reference canvas preserves reference content under overlap while providing an intuitive interface for organizing multiple identities; the subjects remain separate elements on the input canvas, and the model outputs a single personalized and harmonized image. To keep this representation efficient, transparent latent pruning retains only tokens from each subject's non-transparent region, and cross-reference training mitigates copy-paste artifacts by learning to harmonize references sampled from different images. Extensive experiments demonstrate that CanvasComposer achieves coherent generation and strong identity preservation compared to state-of-the-art methods in multi-human personalized image generation. Project page: https://snap-research.github.io/canvascomposer

Gordon Guocheng Qian, Ruihang Zhang, Tsai-Shien Chen +11

Snap Inc., USA · University of Toronto, Canada · University of California, Merced, USA +1

Subject-Driven Image GenerationImage GenerationCompositional Image GenerationIdentity-Preserving Image GenerationConditional Diffusion ModelsMulti-Subject Image Generation
AskPDF