mlsanityData to 7 Oct 2026

PanoDepth: A Two-Stage Approach for Monocular Omnidirectional Depth Estimation

arXiv 2202.01323 ↗First posted 2 Feb 2022cs.CVcited by 9

Not on the map. Why?

Abstract (arXiv)

Omnidirectional 3D information is essential for a wide range of applications such as Virtual Reality, Autonomous Driving, Robotics, etc. In this paper, we propose a novel, model-agnostic, two-stage pipeline for omnidirectional monocular depth estimation. Our proposed framework PanoDepth takes one 360 image as input, produces one or more synthesized views in the first stage, and feeds the original image and the synthesized images into the subsequent stereo matching stage. In the second stage, we propose a differentiable Spherical Warping Layer to handle omnidirectional stereo geometry efficiently and effectively. By utilizing the explicit stereo-based geometric constraints in the stereo matching stage, PanoDepth can generate dense high-quality depth. We conducted extensive experiments and ablation studies to evaluate PanoDepth with both the full pipeline as well as the individual modules in each stage. Our results show that PanoDepth outperforms the state-of-the-art approaches by a large margin for 360 monocular depth estimation.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 48 of its 68 references that are in the map. First posted on arXiv 2 Feb 2022.

Builds on

48 of 68 references.

References · 48

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2017Squeeze-and-Excitation Networkscited by 4,763
  3. 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,154
  4. 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 1,354
  5. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,264
  6. 2016Unsupervised Monocular Depth Estimation with Left-Right Consistencycited by 819
  7. 2018MVSNet: Depth Inference for Unstructured Multi-view Stereocited by 780
  8. 2018Deep Ordinal Regression Network for Monocular Depth Estimationcited by 721
  9. 2018Pyramid Stereo Matching Networkcited by 486
  10. 2014Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecturecited by 483
  11. 2019Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matchingcited by 414
  12. 2017Joint 2D-3D-Semantic Data for Indoor Scene Understandingcited by 404
36 more
  1. 2016Deeper Depth Prediction with Fully Convolutional Residual Networkscited by 402
  2. 2019SynSin: End-to-end View Synthesis from a Single Imagecited by 364
  3. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 359
  4. 2018An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solutioncited by 302
  5. 2018DeepMVS: Learning Multi-view Stereopsiscited by 277
  6. 2019DeepView: View Synthesis with Learned Gradient Descentcited by 253
  7. 2019Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Videocited by 244
  8. 2019Group-wise Correlation Stereo Networkcited by 235
  9. 2018Stereo Magnification: Learning View Synthesis using Multiplane Imagescited by 206
  10. 2017Learning to Predict Indoor Illumination from a Single Imagecited by 160
  11. 2018Revisiting Single Image Depth Estimation: Toward Higher Resolution Maps with Accurate Object Boundariescited by 144
  12. 2018OmniDepth: Dense Depth Estimation for Indoors Spherical Panoramascited by 98
  13. 2016Estimating Depth from Monocular Images as Classification Using Deep Fully Convolutional Residual Networkscited by 91
  14. 2018Layer-structured 3D Scene Inference via View Synthesiscited by 82
  15. 2020Guiding Monocular Depth Estimation Using Depth-Attention Volumecited by 82
  16. 2018PlaneNet: Piece-wise Planar Reconstruction from a Single RGB Imagecited by 77
  17. 2020Monocular Depth Estimation Based On Deep Learning: An Overviewcited by 68
  18. 2019Learning monocular depth estimation infusing traditional stereo knowledgecited by 64
  19. 2018Kernel Transformer Networks for Compact Spherical Convolutioncited by 56
  20. 2019Spherical View Synthesis for Self-Supervised 360 Depth Estimationcited by 51
  21. 2020Learning Stereo from Single Imagescited by 44
  22. 2019360SD-Net: 360deg Stereo Depth Estimation with Learnable Cost Volumecited by 40
  23. 2018Single View Stereo Matchingcited by 39
  24. 2018Deep attention-based classification network for robust depth predictioncited by 36
  25. 2018Eliminating the Blind Spot: Adapting 3D Object Detection and Monocular Depth Estimation to 360deg Panoramic Imagerycited by 34
  26. 2017Deep Learning and Conditional Random Fields-based Depth Estimation and Topographical Reconstruction from Conventional Endoscopycited by 31
  27. 2019Pano Popups: Indoor 3D Reconstruction with a Plane-Aware Networkcited by 25
  28. 2019SweepNet: Wide-baseline Omnidirectional Depth Estimationcited by 24
  29. 2017Learning Spherical Convolution for Fast Features from 360deg Imagerycited by 22
  30. 2017SalNet360: Saliency Maps for omni-directional images with CNNcited by 21
  31. 2019OmniMVS: End-to-End Learning for Omnidirectional Stereo Matchingcited by 21
  32. 2018On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approachcited by 18
  33. 2020ODE-CNN: Omnidirectional Depth Extension Networkscited by 18
  34. 2020Distortion-aware Monocular Depth Estimation for Omnidirectional Imagescited by 17
  35. 2019Mapped Convolutionscited by 14
  36. 2019Convolutions on Spherical Imagescited by 12

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 1; 2024: 1; 2025: 2; 2026: 1.

9 papers.

  1. 2022OmniFusion: 360 Monocular Depth Estimation via Geometry-Aware Fusioncited by 60
  2. 2022Multi-Modal Masked Pre-Training for Monocular Panoramic Depth Completioncited by 23
  3. 2023The RoboDepth Challenge: Methods and Advancements Towards Robust Depth Estimationcited by 13
  4. 2022High-Resolution Depth Estimation for 360-degree Panoramas through Perspective and Panoramic Depth Images Registrationcited by 10
  5. 2025Omni2: Unifying Omnidirectional Image Generation and Editing in an Omni Modelcited by 9
  6. 2025EDM: Equirectangular Projection-Oriented Dense Kernelized Feature Matchingcited by 7
  7. 2022HiMODE: A Hybrid Monocular Omnidirectional Depth Estimation Modelcited by 6
  8. 2024Estimating Depth of Monocular Panoramic Image with Teacher-Student Model Fusing Equirectangular and Spherical Representationscited by 5
  9. 2026Cross360: 360deg Monocular Depth Estimation via Cross Projections Across Scalescited by 0