# Robust Semi-Supervised Monocular Depth Estimation with Reprojected Distances

arXiv: https://arxiv.org/abs/1910.01765 · HTML: https://mlsanity.com/paper/1910.01765/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Oct 2019.

## Built on by (most cited first, 27 of 27)

- [Is Pseudo-Lidar needed for Monocular 3D Object detection?](https://mlsanity.com/paper/2108.06417/) (2021)
- [BinsFormer: Revisiting Adaptive Bins for Monocular Depth Estimation](https://mlsanity.com/paper/2204.00987/) (2022)
- [Multi-Frame Self-Supervised Depth with Transformers](https://mlsanity.com/paper/2204.07616/) (2022)
- [Learning Multiple Dense Prediction Tasks from Partially Annotated Data](https://mlsanity.com/paper/2111.14893/) (2021)
- [Sparse Auxiliary Networks for Unified Monocular Depth Prediction and Completion](https://mlsanity.com/paper/2103.16690/) (2021)
- [Monocular Depth Estimation through Virtual-world Supervision and Real-world SfM Self-Supervision](https://mlsanity.com/paper/2103.12209/) (2021)
- [Multimodal Scale Consistency and Awareness for Monocular Self-Supervised Depth Estimation](https://mlsanity.com/paper/2103.02451/) (2021)
- [Deep Learning-based Depth Estimation Methods from Monocular Image and Videos: A Comprehensive Survey](https://mlsanity.com/paper/2406.19675/) (2024)
- [Distill Any Depth: Distillation Creates a Stronger Monocular Depth Estimator](https://mlsanity.com/paper/2502.19204/) (2025)
- [Advancing Self-supervised Monocular Depth Learning with Sparse LiDAR](https://mlsanity.com/paper/2109.09628/) (2021)
- [Self-supervised Visual-LiDAR Odometry with Flip Consistency](https://mlsanity.com/paper/2101.01322/) (2021)
- [RGB Guided ToF Imaging System: A Survey of Deep Learning-based Methods](https://mlsanity.com/paper/2405.10357/) (2024)
- [LiDARTouch: Monocular metric depth estimation with a few-beam LiDAR](https://mlsanity.com/paper/2109.03569/) (2021)
- [3D-PL: Domain Adaptive Depth Estimation with 3D-aware Pseudo-Labeling](https://mlsanity.com/paper/2209.09231/) (2022)
- [Joint-Task Regularization for Partially Labeled Multi-Task Learning](https://mlsanity.com/paper/2404.01976/) (2024)
- [ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusion](https://mlsanity.com/paper/2407.09303/) (2024)
- [Depth360: Self-supervised Learning for Monocular Depth Estimation using Learnable Camera Distortion Model](https://mlsanity.com/paper/2110.10415/) (2021)
- [Depth Is All You Need for Monocular 3D Detection](https://mlsanity.com/paper/2210.02493/) (2022)
- [Composite Learning for Robust and Effective Dense Predictions](https://mlsanity.com/paper/2210.07239/) (2022)
- [Monocular Depth Estimation for Soft Visuotactile Sensors](https://mlsanity.com/paper/2101.01677/) (2021)
- [Revisiting PatchMatch Multi-View Stereo for Urban 3D Reconstruction](https://mlsanity.com/paper/2207.08439/) (2022)
- [On the Coupling of Depth and Egomotion Networks for Self-Supervised Structure from Motion](https://mlsanity.com/paper/2106.04007/) (2021)
- [Robust Monocular Localization of Drones by Adapting Domain Maps to Depth Prediction Inaccuracies](https://mlsanity.com/paper/2210.15559/) (2022)
- [Dense Road Surface Grip Map Prediction from Multimodal Image Data](https://mlsanity.com/paper/2404.17324/) (2024)
- [Do More With What You Have: Transferring Depth-Scale from Labeled to Unlabeled Domains](https://mlsanity.com/paper/2303.07662/) (2023)
- [MaskingDepth: Masked Consistency Regularization for Semi-supervised Monocular Depth Estimation](https://mlsanity.com/paper/2212.10806/) (2022)
- [View-to-Label: Multi-View Consistency for Self-Supervised 3D Object Detection](https://mlsanity.com/paper/2305.17972/) (2023)

## Abstract (arXiv)

Dense depth estimation from a single image is a key problem in computer vision, with exciting applications in a multitude of robotic tasks. Initially viewed as a direct regression problem, requiring annotated labels as supervision at training time, in the past few years a substantial amount of work has been done in self-supervised depth training based on strong geometric cues, both from stereo cameras and more recently from monocular video sequences. In this paper we investigate how these two approaches (supervised & self-supervised) can be effectively combined, so that a depth model can learn to encode true scale from sparse supervision while achieving high fidelity local accuracy by leveraging geometric cues. To this end, we propose a novel supervised loss term that complements the widely used photometric loss, and show how it can be used to train robust semi-supervised monocular depth estimation models. Furthermore, we evaluate how much supervision is actually necessary to train accurate scale-aware monocular depth models, showing that with our proposed framework, very sparse LiDAR information, with as few as 4 beams (less than 100 valid depth values per image), is enough to achieve results competitive with the current state-of-the-art.