GenMask: Adapting DiT for Segmentation via Direct Mask Generation

• URL: http://arxiv.org/abs/2603.23906v2
• Date: Thu, 26 Mar 2026 04:00:34 GMT
• Title: GenMask: Adapting DiT for Segmentation via Direct Mask Generation
• Abstract summary: We argue that instead of indirect adaptation, segmentation tasks should be trained directly in a generative manner.<n>We introduce timesteps sampling strategy for binary masks that emphasizes extreme noise levels for segmentation and moderate noise for image generation.<n>We present GenMask, a DiT trains to generate black-and-white segmentation masks as well as colorful images in RGB space under the original generative objective.
• Score: 81.54526445834294
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent approaches for segmentation have leveraged pretrained generative models as feature extractors, treating segmentation as a downstream adaptation task via indirect feature retrieval. This implicit use suffers from a fundamental misalignment in representation. It also depends heavily on indirect feature extraction pipelines, which complicate the workflow and limit adaptation. In this paper, we argue that instead of indirect adaptation, segmentation tasks should be trained directly in a generative manner. We identify a key obstacle to this unified formulation: VAE latents of binary masks are sharply distributed, noise robust, and linearly separable, distinct from natural image latents. To bridge this gap, we introduce timesteps sampling strategy for binary masks that emphasizes extreme noise levels for segmentation and moderate noise for image generation, enabling harmonious joint training. We present GenMask, a DiT trains to generate black-and-white segmentation masks as well as colorful images in RGB space under the original generative objective. GenMask preserves the original DiT architecture while removing the need of feature extraction pipelines tailored for segmentation tasks. Empirically, GenMask attains state-of-the-art performance on referring and reasoning segmentation benchmarks and ablations quantify the contribution of each component.

Related papers

• Seg-VAR: Image Segmentation with Visual Autoregressive Modeling [60.79579744943664]
  We propose a novel framework that rethinks segmentation as a conditional autoregressive mask generation problem.<n>This is achieved by replacing the discriminative learning with the latent learning process.<n>Our method incorporates three core components: (1) an image encoder generating latent priors from input images, (2) a spatial-aware seglat (a latent expression of segmentation mask) encoder that maps segmentation masks into discrete latent tokens, and (3) a decoder reconstructing masks from these latents.
  arXiv  Detail & Related papers  (2025-11-16T13:36:19Z)
• GS: Generative Segmentation via Label Diffusion [59.380173266566715]
  Language-driven image segmentation is a fundamental task in vision-language understanding, requiring models to segment regions of an image corresponding to natural language expressions.<n>Recent diffusion models have been introduced to this domain, but existing approaches remain image-centric.<n>We propose GS (Generative Label), a novel framework that formulates segmentation itself as a generative task.<n> Experimental results show that GS significantly outperforms existing discriminative and diffusion-based methods, setting a new state-of-the-art for language-driven segmentation.
  arXiv  Detail & Related papers  (2025-08-27T16:28:15Z)
• SeeDiff: Off-the-Shelf Seeded Mask Generation from Diffusion Models [16.109077391631917]
  We show that cross-attention alone provides very coarse object localization, which however can provide initial seeds.<n>We also observe that a simple-text-guided synthetic image often has a uniform background, which is easier to find correspondences.<n>Our proposed method, dubbed SeeDiff, generates high-quality masks off-the-shelf from Stable Diffusion.
  arXiv  Detail & Related papers  (2025-07-26T05:44:00Z)
• A Simple Latent Diffusion Approach for Panoptic Segmentation and Mask Inpainting [2.7563282688229664]
  This work builds upon Stable Diffusion and proposes a latent diffusion approach for panoptic segmentation. Our training consists of two steps: (1) training a shallow autoencoder to project the segmentation masks to latent space; (2) training a diffusion model to allow image-conditioned sampling in latent space.
  arXiv  Detail & Related papers  (2024-01-18T18:59:19Z)
• SegRefiner: Towards Model-Agnostic Segmentation Refinement with Discrete Diffusion Process [102.18226145874007]
  We propose a model-agnostic solution called SegRefiner to enhance the quality of object masks produced by different segmentation models. SegRefiner takes coarse masks as inputs and refines them using a discrete diffusion process. It consistently improves both the segmentation metrics and boundary metrics across different types of coarse masks.
  arXiv  Detail & Related papers  (2023-12-19T18:53:47Z)
• DFormer: Diffusion-guided Transformer for Universal Image Segmentation [86.73405604947459]
  The proposed DFormer views universal image segmentation task as a denoising process using a diffusion model. At inference, our DFormer directly predicts the masks and corresponding categories from a set of randomly-generated masks. Our DFormer outperforms the recent diffusion-based panoptic segmentation method Pix2Seq-D with a gain of 3.6% on MS COCO val 2017 set.
  arXiv  Detail & Related papers  (2023-06-06T06:33:32Z)
• Cross-modal tumor segmentation using generative blending augmentation and self training [1.6440045168835438]
  We propose a cross-modal segmentation method based on conventional image synthesis boosted by a new data augmentation technique. Generative Blending Augmentation (GBA) learns representative generative features from a single training image to realistically diversify tumor appearances. The proposed solution ranked first for vestibular schwannoma (VS) segmentation during the validation and test phases of the MICCAI CrossMoDA 2022 challenge.
  arXiv  Detail & Related papers  (2023-04-04T11:01:46Z)
• Complementary Random Masking for RGB-Thermal Semantic Segmentation [63.93784265195356]
  RGB-thermal semantic segmentation is a potential solution to achieve reliable semantic scene understanding in adverse weather and lighting conditions. This paper proposes 1) a complementary random masking strategy of RGB-T images and 2) self-distillation loss between clean and masked input modalities. We achieve state-of-the-art performance over three RGB-T semantic segmentation benchmarks.
  arXiv  Detail & Related papers  (2023-03-30T13:57:21Z)
• Generative Semantic Segmentation [40.57488730457299]
  We present a generative learning approach for semantic segmentation. Uniquely, we cast semantic segmentation as an image-conditioned mask generation problem. Experiments show that our GSS can perform competitively to prior art alternatives in the standard semantic segmentation setting.
  arXiv  Detail & Related papers  (2023-03-20T17:55:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.