Seg2Any: Open-set Segmentation-Mask-to-Image Generation with Precise Shape and Semantic Control

TL;DR

Seg2Any introduces a novel segmentation-mask-to-image generation framework that achieves precise spatial layout, semantic, and shape control by decoupling conditions and employing advanced multimodal diffusion transformers, supported by a large-scale dataset.

Contribution

The paper presents Seg2Any, a new S2I framework with decoupled semantic and shape conditions, attribute isolation, and a large dataset for open-set generation, advancing spatial and attribute control in image synthesis.

Findings

State-of-the-art performance on S2I benchmarks.

Effective semantic and shape consistency in generated images.

Robust attribute control in multi-entity scenarios.

Abstract

Despite recent advances in diffusion models, top-tier text-to-image (T2I) models still struggle to achieve precise spatial layout control, i.e. accurately generating entities with specified attributes and locations. Segmentation-mask-to-image (S2I) generation has emerged as a promising solution by incorporating pixel-level spatial guidance and regional text prompts. However, existing S2I methods fail to simultaneously ensure semantic consistency and shape consistency. To address these challenges, we propose Seg2Any, a novel S2I framework built upon advanced multimodal diffusion transformers (e.g. FLUX). First, to achieve both semantic and shape consistency, we decouple segmentation mask conditions into regional semantic and high-frequency shape components. The regional semantic condition is introduced by a Semantic Alignment Attention Mask, ensuring that generated entities adhere to their assigned text prompts. The high-frequency shape condition, representing entity boundaries, is encoded as an Entity Contour Map and then introduced as an additional modality via multi-modal attention to guide image spatial structure. Second, to prevent attribute leakage across entities in multi-entity scenarios, we introduce an Attribute Isolation Attention Mask mechanism, which constrains each entity's image tokens to attend exclusively to themselves during image self-attention. To support open-set S2I generation, we construct SACap-1M, a large-scale dataset containing 1 million images with 5.9 million segmented entities and detailed regional captions, along with a SACap-Eval benchmark for comprehensive S2I evaluation. Extensive experiments demonstrate that Seg2Any achieves state-of-the-art performance on both open-set and closed-set S2I benchmarks, particularly in fine-grained spatial and attribute control of entities.