从RGB生成到密集场读出：基于文本到图像模型的像素空间密集预测

摘要

大规模文本到图像模型因其RGB生成预训练能学习丰富的语义、结构与几何先验，成为密集预测任务中极具吸引力的主干架构。现有生成式与编辑方法通过将密集预测转化为目标生成来复用这些先验：深度、法线、透明度蒙版、分割掩码、热力图等标注被编码至RGB预训练的VAE潜在空间，再解码为类图像目标。我们认为，这种做法继承了比密集预测所需更多的生成式输出接口——与RGB合成不同，密集预测要求在同一图像平面上输出像素精确、任务原生的场，而非渲染新的RGB内容。我们的关键观察是，预训练的DiT已通过图像平面上的"块到令牌到块"网格结构组织RGB输入，因此每个令牌索引一个固定的输出块，其通道可承载任务原生量而非RGB外观。我们将其实现为ReChannel：保留用于DiT输入分布的VAE编码器，但舍弃目标侧解码器；通过任务LoRA适配冻结的DiT，并通过共享的令牌局部线性头（约33K参数，无空间混合）将每个令牌映射至其p×p×K_t像素空间块。基于FLUX-Klein，我们在六个密集预测任务及十余个基准上评估了该方法。这一极简接口在无三元图抠图、KITTI深度和指代分割上创下新最优结果，并在法线、显著性、姿态估计上保持竞争力。在匹配的4B参数设定下，其准确率更高且速度比"编辑+潜在解码"方案快2.48倍——密集感知可从生成式预训练中获益，而无需继承其输出接口。

English

Large-scale text-to-image models are attractive backbones for dense prediction because RGB generation pretraining learns rich semantic, structural, and geometric priors. Existing generative and editing approaches reuse these priors by casting dense prediction as target generation: annotations such as depth, normals, alpha mattes, masks, and heatmaps are encoded into an RGB-trained VAE latent space and decoded back as image-like targets. We argue this inherits more of the generative output interface than dense prediction requires: unlike RGB synthesis, dense prediction asks for pixel-correct, task-native fields on the same image plane, not new RGB content to be rendered. Our key observation is that a pretrained DiT already organizes RGB inputs through a patch-to-token-to-patch lattice on the image plane, so each token indexes a fixed output patch whose channels can carry task-native quantities instead of RGB appearance. We instantiate this as ReChannel: we keep the VAE encoder for the DiT's input distribution but drop the target-side decoder, adapt the frozen DiT with task LoRA, and map each token to its p x p x K_t pixel-space patch through a shared token-local linear head--about 33K parameters, no spatial mixing. Using FLUX-Klein, we evaluate on six dense prediction tasks and over a dozen benchmarks. This minimal interface sets new state-of-the-art on trimap-free matting, KITTI depth, and referring segmentation, and stays competitive on normals, saliency, and pose. In a matched 4B setting it is more accurate and 2.48x faster than an edit-plus-latent-decode counterpart--dense perception can benefit from generative pretraining without inheriting its output interface.