arXiv:2603. 12261v2 Announce Type: replace-cross Abstract: Text-to-image generation models have advanced rapidly, yet achieving fine-grained control over generated images remains difficult, largely due to limited understanding of how semantic information is encoded.
Paint-Anything introduces a unified hex-prompt interface that allows users to specify any 24‑bit hex color for both image generation and editing. The method trains on a new Paint‑500K dataset created from real images with object grounding, perceptual color labeling, and editing‑pair synthesis, and supplements this with pure‑color anchors to address shadow‑induced color inaccuracies. Evaluated on the newly proposed Any Color Benchmark (ACBench), Paint‑Anything achieves significant improvements over the base FLUX.2‑4B model, boosting T2I and editing scores by 85.3 % and 28.3 % respectively, and outperforms competing methods on the CompColor metric.
arXiv:2610.09450v1 Announce Type: new Abstract: Pixel-space diffusion models avoid the lossy VAE of latent models, which suggests an advantage on downstream tasks where fine-grained detail matters. W...
Score Distillation Sampling (SDS) enables text-to-3D generation by optimizing rendered images with a pretrained diffusion prior, but latent SDS often produces structured color artifacts and high-frequency texture noise. We identify a failure mode of latent SDS caused by VAE-induced pixel drift: the optimized image can move along pixel-space directions that are weakly constrained by the VAE encoder, so its latent representation remains clean and semantically meaningful while the image itself accumulates visible artifacts.
DiDE is a training‑free framework that enables independent control of color and texture when stylizing 3D assets generated by rectified flow‑based image‑to‑3D models. It leverages the overcomplete latent space of these models, partitioning channels so that texture occupies a small subset while color is encoded in a free subspace. Experiments on the newly collected Disen3D‑Bench benchmark demonstrate that DiDE outperforms existing 2D and 3D stylization methods in color fidelity, texture transfer, and content preservation.
arXiv:2608. 10798v1 Announce Type: cross Abstract: Most image colorization systems operate in $Lab$ space by predicting chroma ($ab$) while preserving an input-derived luminance channel ($L$).
The paper introduces Colorist, a data‑augmentation method that uses classical statistical color matching to generate domain‑shifted medical images. By applying global mean‑standard‑deviation matching in RGB space, Colorist creates structurally intact variations without neural networks, outperforming deep generative models in fidelity and color alignment. Across multiple medical imaging datasets, it boosts balanced accuracy by up to 9% over state‑of‑the‑art domain‑generalization regularizers and 13% over no augmentation, while reducing computational cost and preserving anatomical structure.
arXiv:2607. 13647v1 Announce Type: cross Abstract: Do vision models see colors the way humans do?
arXiv:2608.21847v1 Announce Type: new Abstract: Low-light image enhancement (LLIE) must correct ambiguous exposure without overwriting structure already supported by the input. Generative transport c...
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.
The paper introduces a method that applies physically based rendering (PBR) within the latent space of variational autoencoders used in image diffusion models. By modifying the rendering equation and using a differentiable renderer, the authors can generate latent maps that guide content creation with physically accurate lighting. The approach is trained on a single rendered image and then shown to generalize to changes in scene geometry, lighting, and camera viewpoint.
ChromaGS is a real‑time, language‑guided method for color editing of animatable 3D Gaussian head avatars. It augments each Gaussian primitive with learned soft assignments to semantic regions and decomposes colors into region‑level base colors and Gaussian‑level residuals, enabling coherent color transfer while preserving fine details. A two‑stage language pipeline translates natural‑language instructions into target colors, supporting both absolute and relative adjustments without requiring retraining.