ART: Anonymous Region Transformer for Variable Multi-Layer Transparent Image Generation

TL;DR

ART introduces Anonymous Region Transformer to enable variable multi-layer transparent image generation driven by a global prompt and anonymous region layouts. The approach decomposes into a Multi-layer Transparent Image Autoencoder, the ART diffusion backbone with layout-conditioned 3D RoPE, and an Anonymous Region Layout Planner, trained on the MLTD dataset (~1M designs). By performing region-wise cropping and joint diffusion across layers, ART achieves 50+ layers with improved coherence, efficiency, and cross-layer harmonization, outperforming prior methods in both quantitative metrics and user studies. This work enables scalable, interactive, layer-wise content creation and lays a foundation for future improvements in semantic labeling, aesthetics, and human-in-the-loop editing.

Abstract

Multi-layer image generation is a fundamental task that enables users to isolate, select, and edit specific image layers, thereby revolutionizing interactions with generative models. In this paper, we introduce the Anonymous Region Transformer (ART), which facilitates the direct generation of variable multi-layer transparent images based on a global text prompt and an anonymous region layout. Inspired by Schema theory suggests that knowledge is organized in frameworks (schemas) that enable people to interpret and learn from new information by linking it to prior knowledge.}, this anonymous region layout allows the generative model to autonomously determine which set of visual tokens should align with which text tokens, which is in contrast to the previously dominant semantic layout for the image generation task. In addition, the layer-wise region crop mechanism, which only selects the visual tokens belonging to each anonymous region, significantly reduces attention computation costs and enables the efficient generation of images with numerous distinct layers (e.g., 50+). When compared to the full attention approach, our method is over 12 times faster and exhibits fewer layer conflicts. Furthermore, we propose a high-quality multi-layer transparent image autoencoder that supports the direct encoding and decoding of the transparency of variable multi-layer images in a joint manner. By enabling precise control and scalable layer generation, ART establishes a new paradigm for interactive content creation.

Figures (14)

• Figure 1: Semantic Layout vs. Anonymous Region Layout. The conventional semantic layout requires specifying what objects to generate in each given region, whereas our anonymous region layout only identifies where the important regions are. People can leverage the prior knowledge, activated by the global prompt, to intuitively infer the semantic label of each anonymous region. The generative model also learns to harness this capability and autonomously determine what to generate in each region.
• Figure 1: Conflicts presented in Semantic Layout based Results: We display the composed entire image in the 1st column, the reference image in the 2nd column, and the semantic layout in the 3rd column. The conflicted regions are marked with red bounding boxes in both the composed entire images and the reference images. We visualize the attention maps between semantic regions, region-wise prompts, and the global reference images.
• Figure 2: Visual planning capability of our Anonymous Region Transformer. We visualize the averaged attention maps of all visual tokens within the same anonymous region (as Query) attending to the entities within the global prompt text tokens (as Key and Value). These attention maps reveal that each anonymous region assigns the majority of attention weights to one of the major objects identified in the given text prompt.
• Figure 3: ART vs. previous SOTA in multi-layer transparent image generation: user study results across different domains. ART significantly outperforms LayerDiffuse zhang2024transparent in the photorealistic domain and COLE jia2023cole in the graphic-design domain across multiple aspects.
• Figure 3: Generated Result with 40 transparent image layers. Top-left: Generated Merged Image; Top-Right: Generated Transparent Layers; Bottom-left: Anonymous Region Layout; Bottom-right: Global Prompt.