PixelART：无需潜变量或文本到图像预训练的图像到图层分解
PixelART: Image-to-Layer Decomposition without Latents or Text-to-Image Pretraining
- Tsinghua University（清华大学）
- Canva Research（Canva研究院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
PixelART提出从零训练的像素空间整流流Transformer，无需潜变量或T2I预训练，实现高效图像到RGBA图层分解，参数、延迟和内存大幅降低，性能达到最先进水平。
AI中文摘要：
图像到图层分解将扁平图像转换为可编辑的RGBA图层，从而在设计工作流中实现元素级编辑。现有的基于扩散的系统通常改编大型预训练的文本到图像（T2I）模型，并引入RGBA自编码器或可变图层架构模块。我们重新审视这一设计选择，并质疑图层分解是否真正需要这些重量级组件。我们提出PixelART，一种从零训练的像素空间整流流Transformer，用于图像到图层（I2L）分解。PixelART使用单流多模态扩散Transformer直接对区域RGBA像素块进行去噪，避免了RGBA-VAE、预训练的T2I骨干网络和特定于图层的解码器。我们识别出该任务的一个关键特性：高噪声时间步决定图层分配和粗略图层组织，而低噪声时间步主要细化颜色、透明度、纹理和边界。基于这一观察，我们提出终端增强的时间步采样策略，以增加高噪声图层分配区域的训练覆盖。在400万多层设计模板上训练，PixelART在Design-Multi-Layer-Bench和LICA上实现了最先进的图层分解和合成重建，与最近的Qwen-Image-Layered模型相比，参数减少超过80%，延迟降低98%，内存降低85%。消融实验表明，像素空间x预测、终端增强的时间步采样以及数据/模型缩放至关重要，而T2I预训练对I2L任务带来的收益微乎其微。
英文摘要：
Image-to-layer decomposition converts a flattened image into editable RGBA layers, enabling element-level editing in design workflows. Existing diffusion-based systems typically adapt large pretrained text-to-image (T2I) models and introduce RGBA autoencoders or variable-layer architectural modules. We revisit this design choice and ask whether layer decomposition truly requires these heavyweight components. We introduce PixelART, a pixel-space rectified-flow Transformer trained from scratch for image-to-layer (I2L) decomposition. PixelART directly denoises regional RGBA pixel patches using a single-stream multi-modal diffusion Transformer, avoiding RGBA-VAEs, pretrained T2I backbones, and layer-specific decoders. We identify a key property of the task: high-noise timesteps determine layer assignment and coarse layer organization, while low-noise timesteps mainly refine color, alpha, texture, and boundaries. Based on this observation, we propose a terminal-boosted timestep sampling strategy to increase training coverage in the high-noise layer assignment regime. Trained on 4M multi-layer design templates, PixelART achieves state-of-the-art layer decomposition and composite reconstruction on Design-Multi-Layer-Bench and LICA with over 80% fewer parameters, 98% lower latency, and 85% lower memory than the recent Qwen-Image-Layered model. Ablation experiments show that pixel-space $\mathbf{x}$-prediction, terminal-boosted timestep sampling, and data/model scaling are critical, while T2I pretraining brings marginal benefits to the I2L task.
