# From Editor to Dense Geometry Estimator

arXiv: https://arxiv.org/abs/2509.04338 · HTML: https://mlsanity.com/paper/2509.04338/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models.
- Builds on 61 of its 90 references that are in the map.
- First posted on arXiv 4 Sep 2025.

## Placed in

- Diffusion models / [3D, point cloud and scene generation](https://mlsanity.com/diffusion/3d-point-cloud-and-scene-generation/)

## Builds on (most cited first, 30 of 61)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [InstructPix2Pix: Learning to Follow Image Editing Instructions](https://mlsanity.com/paper/2211.09800/) (2022)
- [Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://mlsanity.com/paper/1511.06434/) (2015)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer](https://mlsanity.com/paper/1907.01341/) (2019)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://mlsanity.com/paper/2401.10891/) (2024)
- [Depth Map Prediction from a Single Image using a Multi-Scale Deep Network](https://mlsanity.com/paper/1406.2283/) (2014)
- [Alias-Free Generative Adversarial Networks](https://mlsanity.com/paper/2106.12423/) (2021)
- [PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis](https://mlsanity.com/paper/2310.00426/) (2023)
- [AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networks](https://mlsanity.com/paper/1711.10485/) (2017)
- [Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding](https://mlsanity.com/paper/2011.02523/) (2020)
- [StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks](https://mlsanity.com/paper/1612.03242/) (2016)
- [Mean Flows for One-step Generative Modeling](https://mlsanity.com/paper/2505.13447/) (2025)
- [Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecture](https://mlsanity.com/paper/1411.4734/) (2014)
- [Step1X-Edit: A Practical Framework for General Image Editing](https://mlsanity.com/paper/2504.17761/) (2025)
- [Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimation](https://mlsanity.com/paper/2404.15506/) (2024)

## Built on by (most cited first, 24 of 24)

- [Pixel-Perfect Depth with Semantics-Prompted Diffusion Transformers](https://mlsanity.com/paper/2510.07316/) (2025)
- [What if Agents Could Imagine? Reinforcing Open-Vocabulary HOI Comprehension through Generation](https://mlsanity.com/paper/2602.11499/) (2026)
- [From Scale to Speed: Adaptive Test-Time Scaling for Image Editing](https://mlsanity.com/paper/2603.00141/) (2026)
- [Semantic Context Matters: Improving Conditioning for Autoregressive Models](https://mlsanity.com/paper/2511.14063/) (2025)
- [Eevee: Towards Close-up High-resolution Video-based Virtual Try-on](https://mlsanity.com/paper/2511.18957/) (2025)
- [PaCo-RL: Advancing Reinforcement Learning for Consistent Image Generation with Pairwise Reward Modeling](https://mlsanity.com/paper/2512.04784/) (2025)
- [Vision as Unified Multimodal Generation](https://mlsanity.com/paper/2607.06560/) (2026)
- [CaC: Advancing Video Reward Models via Hierarchical Spatiotemporal Concentrating](https://mlsanity.com/paper/2605.11723/) (2026)
- [Large Depth Completion Model from Sparse Observations](https://mlsanity.com/paper/2605.30115/) (2026)
- [End-to-End Multi-Modal Diffusion Mamba](https://mlsanity.com/paper/2510.13253/) (2025)
- [FMPose3D: monocular 3D pose estimation via flow matching](https://mlsanity.com/paper/2602.05755/) (2026)
- [Elucidating the SNR-t Bias of Diffusion Probabilistic Models](https://mlsanity.com/paper/2604.16044/) (2026)
- [Towards Consistent Video Geometry Estimation](https://mlsanity.com/paper/2605.30060/) (2026)
- [DetailAnywhere: Fashion Detail Generation via Cross-Modal Feature Alignment Distillation](https://mlsanity.com/paper/2607.02220/) (2026)
- [Pixel-Perfect Visual Geometry Estimation](https://mlsanity.com/paper/2601.05246/) (2026)
- [IndusAgent: Reinforcing Open-Vocabulary Industrial Anomaly Detection with Agentic Tools](https://mlsanity.com/paper/2605.20682/) (2026)
- [From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models](https://mlsanity.com/paper/2607.06553/) (2026)
- [RePer-360: Releasing Perspective Priors for 360^circ Depth Estimation via Self-Modulation](https://mlsanity.com/paper/2603.05999/) (2026)
- [GeoStereo: A Unified Stereo Geometry Estimation Framework for Disparity and Surface Normal](https://mlsanity.com/paper/2607.24024/) (2026)
- [RenderMatte: Exact-Alpha Rendering and Group-Relative Alignment for Image Matting](https://mlsanity.com/paper/2608.08487/) (2026)
- [TransNormal-2: Geometry-Grounded Rectified Flow with Edge-Aware Decoding for Precise Normal Estimation](https://mlsanity.com/paper/2609.06665/) (2026)
- [VepAgent: Bridging Causal-Transition via Tool-Augmented Reinforcement Learning for Video Event Prediction](https://mlsanity.com/paper/2610.06293/) (2026)
- [X-Ragent: Tool-Augmented Reinforcement Learning for Evidence-Grounded Chest X-Ray Reasoning](https://mlsanity.com/paper/2610.16811/) (2026)
- [OpenDepth: Repurposing Image Editing Priors for Robust Open-World Monocular Depth Estimation](https://mlsanity.com/paper/2610.17610/) (2026)

## Abstract (arXiv)

Leveraging visual priors from pre-trained text-to-image (T2I) generative models has shown success in dense prediction. However, dense prediction is inherently an image-to-image task, suggesting that image editing models, rather than T2I generative models, may be a more suitable foundation for fine-tuning. Motivated by this, we conduct a systematic analysis of the fine-tuning behaviors of both editors and generators for dense geometry estimation. Our findings show that editing models possess inherent structural priors, which enable them to converge more stably by ``refining" their innate features, and ultimately achieve higher performance than their generative counterparts. Based on these findings, we introduce FE2E, a framework that pioneeringly adapts an advanced editing model based on Diffusion Transformer (DiT) architecture for dense geometry prediction. Specifically, to tailor the editor for this deterministic task, we reformulate the editor's original flow matching loss into the ``consistent velocity" training objective. And we use logarithmic quantization to resolve the precision conflict between the editor's native BFloat16 format and the high precision demand of our tasks. Additionally, we leverage the DiT's global attention for a cost-free joint estimation of depth and normals in a single forward pass, enabling their supervisory signals to mutually enhance each other. Without scaling up the training data, FE2E achieves impressive performance improvements in zero-shot monocular depth and normal estimation across multiple datasets. Notably, it achieves over 35% performance gains on the ETH3D dataset and outperforms the DepthAnything series, which is trained on 100× data. The project page can be accessed hrefhttps://amap-ml.github.io/FE2E/here.