Modality Forcing for Scalable Spatial Generation

• URL: http://arxiv.org/abs/2606.13676v1
• Date: Thu, 11 Jun 2026 17:59:45 GMT
• Title: Modality Forcing for Scalable Spatial Generation
• Abstract summary: Text-to-image (T2I) models contain rich spatial priors.<n>Prior works adapt T2I models to leverage this prior for depth prediction, but they require dense depth data and involve complex recipes.<n>We propose Modality Forcing, a simple, scalable post-training recipe for joint image-depth generation using a single DiT trained on sparse depth data.
• Score: 54.04539566839143
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Text-to-image (T2I) models contain rich spatial priors. Synthesizing photorealistic, cluttered scenes requires an understanding of geometry, including perspective and relative scale. Prior works adapt T2I models to leverage this prior for depth prediction, but they require dense depth data and involve complex recipes. We propose Modality Forcing, a simple, scalable post-training recipe for joint image-depth generation using a single DiT trained on sparse depth data. Modality Forcing enables conditional and joint generation of image and depth in any permutation by assigning separate noise levels per modality. Per-modality decoders let us train on sparse, real-world depth and achieve strong, generalizable depth prediction. We further show that Modality Forcing inherits the scalability of T2I pre-training: by training a set of T2I models from scratch (370M to 3.3B parameters), we find that larger models trained on more image data produce more accurate depth. Our strongest model is competitive with state-of-the-art monocular depth estimators and reduces AbsRel by 57% relative to existing joint image-depth generative models. These results provide strong evidence that image generation is a scalable pre-training objective for spatial perception. https://modality-forcing.github.io/

Related papers

• BRIDGE -- Building Reinforcement-Learning Depth-to-Image Data Generation Engine for Monocular Depth Estimation [17.554501937884172]
  BRIDGE is an RL-optimized depth-to-image (D2I) generation framework.<n>It synthesizes over 20M realistic and geometrically accurate RGB images, each intrinsically paired with its ground truth depth.<n>We train our depth estimation model on this dataset, employing a hybrid supervision strategy.
  arXiv  Detail & Related papers  (2025-09-29T17:19:45Z)
• From Editor to Dense Geometry Estimator [77.21804448599009]
  We introduce textbfFE2E, a framework that adapts an advanced editing model based on Diffusion Transformer (DiT) architecture for dense geometry prediction.<n>FE2E achieves over 35% performance gains on the ETH3D dataset and outperforms the DepthAnything series, which is trained on 100$times$ data.
  arXiv  Detail & Related papers  (2025-09-04T15:58:50Z)
• Propagating Sparse Depth via Depth Foundation Model for Out-of-Distribution Depth Completion [33.854696587141355]
  We propose a novel depth completion framework that leverages depth foundation models to attain remarkable robustness without large-scale training.<n>Specifically, we leverage a depth foundation model to extract environmental cues, including structural and semantic context, from RGB images to guide the propagation of sparse depth information into missing regions.<n>Our framework performs remarkably well in the OOD scenarios and outperforms existing state-of-the-art depth completion methods.
  arXiv  Detail & Related papers  (2025-08-07T02:38:24Z)
• Distilling Monocular Foundation Model for Fine-grained Depth Completion [17.603217168518356]
  We propose a two-stage knowledge distillation framework to provide dense supervision for depth completion.<n>In the first stage, we generate diverse training data from natural images, which distills geometric knowledge to depth completion.<n>In the second stage, we employ a scale- and shift-invariant loss to learn real-world scales when fine-tuning on real-world datasets.
  arXiv  Detail & Related papers  (2025-03-21T09:34:01Z)
• DistillNeRF: Perceiving 3D Scenes from Single-Glance Images by Distilling Neural Fields and Foundation Model Features [65.8738034806085]
  DistillNeRF is a self-supervised learning framework for understanding 3D environments in autonomous driving scenes. Our method is a generalizable feedforward model that predicts a rich neural scene representation from sparse, single-frame multi-view camera inputs.
  arXiv  Detail & Related papers  (2024-06-17T21:15:13Z)
• Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data [87.61900472933523]
  This work presents Depth Anything, a highly practical solution for robust monocular depth estimation. We scale up the dataset by designing a data engine to collect and automatically annotate large-scale unlabeled data. We evaluate its zero-shot capabilities extensively, including six public datasets and randomly captured photos.
  arXiv  Detail & Related papers  (2024-01-19T18:59:52Z)
• Towards Accurate Reconstruction of 3D Scene Shape from A Single Monocular Image [91.71077190961688]
  We propose a two-stage framework that first predicts depth up to an unknown scale and shift from a single monocular image. We then exploits 3D point cloud data to predict the depth shift and the camera's focal length that allow us to recover 3D scene shapes. We test our depth model on nine unseen datasets and achieve state-of-the-art performance on zero-shot evaluation.
  arXiv  Detail & Related papers  (2022-08-28T16:20:14Z)
• Adaptive Context-Aware Multi-Modal Network for Depth Completion [107.15344488719322]
  We propose to adopt the graph propagation to capture the observed spatial contexts. We then apply the attention mechanism on the propagation, which encourages the network to model the contextual information adaptively. Finally, we introduce the symmetric gated fusion strategy to exploit the extracted multi-modal features effectively. Our model, named Adaptive Context-Aware Multi-Modal Network (ACMNet), achieves the state-of-the-art performance on two benchmarks.
  arXiv  Detail & Related papers  (2020-08-25T06:00:06Z)

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.