From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models
:::info Stub — Full Engineering Breakdown Coming This paper was featured on Hugging Face Daily Papers on 2026-07-09 with 10 upvotes. A full breakdown with production viability rating, implementation notes, and honest limitations is being written. Subscribe to AI Letters → :::
| Authors | Zanyi Wang et al. |
| Year | 2026 |
| HF Upvotes | 10 |
| arXiv | 2607.06553 |
| Download | |
| HF Page | View on Hugging Face |
Abstract
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.
Engineering Breakdown
The Problem
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.
The Approach
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.
Key Results
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.
Research Areas
This paper contributes to the following areas of AI/ML engineering:
- Machine learning
- Deep learning
- Neural networks
- Model optimization
- AI systems
- Generation
:::tip Subscribe Get weekly breakdowns of papers like this in AI Letters - the newsletter for engineers building production AI systems. :::