mlsanityData to 14 Oct 2026

The Midas Touch for Metric Depth

arXiv 2605.11578 ↗First posted 12 May 2026cs.CVcited by 2

Not on the map. Why?

Abstract (arXiv)

Recent advances have markedly improved the cross-scene generalization of relative depth estimation, yet its practical applicability remains limited by the absence of metric scale, local inconsistencies, and low computational efficiency. To address these issues, we present emphMidas Touch for Depth (MTD), a mathematically interpretable approach that converts relative depth into metric depth using only extremely sparse 3D data. To eliminate local scale inconsistencies, it applies a segment-wise recovery strategy via sparse graph optimization, followed by a pixel-wise refinement strategy using a discontinuity-aware geodesic cost. MTD exhibits strong generalization and achieves substantial accuracy improvements over previous depth completion and depth estimation methods. Moreover, its lightweight, plug-and-play design facilitates deployment and integration on diverse downstream 3D tasks. Project page is available at https://mias.group/MTD.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 64 of its 81 references that are in the map. First posted on arXiv 12 May 2026.

Builds on

64 of 81 references.

References · 64

  1. 2023Segment Anythingcited by 8,234
  2. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  3. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  4. 2021Vision Transformers for Dense Predictioncited by 1,803
  5. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,584
  6. 2025VGGT: Visual Geometry Grounded Transformercited by 1,539
  7. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,495
  8. 2024Depth Anything V2cited by 1,422
  9. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,360
  10. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,274
  11. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 984
  12. 2016Unsupervised Monocular Depth Estimation with Left-Right Consistencycited by 820
38 more
  1. 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 687
  2. 2020Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understandingcited by 604
  3. 2020TartanAir: A Dataset to Push the Limits of Visual SLAMcited by 506
  4. 20193D Packing for Self-Supervised Monocular Depth Estimationcited by 439
  5. 2024Depth Pro: Sharp Monocular Metric Depth in Less Than a Secondcited by 404
  6. 2024Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimationcited by 395
  7. 2024UniDepth: Universal Monocular Metric Depth Estimationcited by 382
  8. 2023Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Imagecited by 365
  9. 2020Virtual KITTI 2cited by 353
  10. 2021RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matchingcited by 353
  11. 2023Faster Segment Anything: Towards Lightweight SAM for Mobile Applicationscited by 352
  12. 2021Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scanscited by 336
  13. 2023Repurposing Diffusion-Based Image Generators for Monocular Depth Estimationcited by 327
  14. 2024GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Imagecited by 286
  15. 2019DIODE: A Dense Indoor and Outdoor DEpth Datasetcited by 240
  16. 2022TinyViT: Fast Pretraining Distillation for Small Vision Transformerscited by 234
  17. 2020Learning to Recover 3D Scene Shape from a Single Imagecited by 216
  18. 2020Non-Local Spatial Propagation Network for Depth Completioncited by 195
  19. 2023Iterative Geometry Encoding Volume for Stereo Matchingcited by 190
  20. 2023EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anythingcited by 188
  21. 2023MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimationcited by 174
  22. 2025UniDepthV2: Universal Monocular Metric Depth Estimation Made Simplercited by 173
  23. 2023EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attentioncited by 172
  24. 2024Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Predictioncited by 166
  25. 2018Depth Estimation via Affinity Learned with Convolutional Spatial Propagation Networkcited by 110
  26. 2019Domain-invariant Stereo Matching Networkscited by 109
  27. 2022Lite-Mono: A Lightweight CNN and Transformer Architecture for Self-Supervised Monocular Depth Estimationcited by 100
  28. 2024Selective-Stereo: Adaptive Frequency Information Selection for Stereo Matchingcited by 99
  29. 2023CompletionFormer: Depth Completion with Convolutions and Vision Transformerscited by 93
  30. 2019Unsupervised Depth Completion from Visual Inertial Odometrycited by 89
  31. 2024Prompting Depth Anything for 4K Resolution Accurate Metric Depth Estimationcited by 74
  32. 2022Dynamic Spatial Propagation Network for Depth Completioncited by 71
  33. 2023Behind the Scenes: Density Fields for Single View Reconstructioncited by 63
  34. 2020DiverseDepth: Affine-invariant Depth Prediction Using Diverse Datacited by 61
  35. 2024Bilateral Propagation Network for Depth Completioncited by 60
  36. 2024IGEV++: Iterative Multi-range Geometry Encoding Volumes for Stereo Matchingcited by 59
  37. 2022Hierarchical Normalization for Robust Monocular Depth Estimationcited by 52
  38. 2024Marigold-DC: Zero-Shot Monocular Depth Completion with Guided Diffusioncited by 50

50 most cited of 64 All 64 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Boosting Metric Depth Completion via Training-Free Adaptive Response Geometrycited by 0
  2. 2026The Midas Touch for Metric Depth: Interpretable Metricization for Generalizable Depth Perceptioncited by 0