SIGMA-GEN: Structure and Identity Guided Multi-subject Assembly for Image Generation
It addresses the problem of generating images with multiple subjects while preserving their identities for applications in creative and synthetic media, representing a novel advancement rather than incremental.
The paper tackles multi-identity preserving image generation by introducing SIGMA-GEN, a framework that enables single-pass generation with structural and spatial constraints, achieving state-of-the-art performance in identity preservation, quality, and speed.
We present SIGMA-GEN, a unified framework for multi-identity preserving image generation. Unlike prior approaches, SIGMA-GEN is the first to enable single-pass multi-subject identity-preserved generation guided by both structural and spatial constraints. A key strength of our method is its ability to support user guidance at various levels of precision -- from coarse 2D or 3D boxes to pixel-level segmentations and depth -- with a single model. To enable this, we introduce SIGMA-SET27K, a novel synthetic dataset that provides identity, structure, and spatial information for over 100k unique subjects across 27k images. Through extensive evaluation we demonstrate that SIGMA-GEN achieves state-of-the-art performance in identity preservation, image generation quality, and speed. Code and visualizations at https://oindrilasaha.github.io/SIGMA-Gen/