Monocular Depth Estimation via Listwise Ranking using the Plackett-Luce Model

Lienen, Julian; Hüllermeier, Eyke; Ewerth, Ralph; Nommensen, Nils

doi:10.1109/cvpr46437.2021.01436

Cited by 12 publications

(6 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In relative depth estimation, the scale-invariant loss [13] and its variants [33,34,40,48] have been used to cope with the scale ambiguity of depth labels. Recently, listwise ranking [35], instead of pairwise ranking, and depth normalization [23] have been considered.…”

Section: Relative Depth Estimationmentioning

confidence: 99%

“…There have been some attempts for this conversion. For instance, relative depths were fitted to metric depths directly using least-squares in [35,40], and a relative depth estimator was fine-tuned to a metric depth estimator in [39]. Also, relative and metric depths were jointly learned through depth map decomposition in [23].…”

Section: Relative-to-metric Depth Conversionmentioning

confidence: 99%

“…However, [35,40] require a large depth-labeled dataset for fitting. Also, [23,39] fine-tune a pre-trained network to estimate metric depth maps for a single camera only.…”

Section: Relative-to-metric Depth Conversionmentioning

confidence: 99%

“…In particular, joint depth estimation of both indoor and outdoor scenes has been considered incompatible for a single network. To overcome this issue, relative depth estimation, which predicts the relative depth order among pixels instead of absolute (or metric) depths, has been studied [6,35,51,58]. Since the depth order is camera-invariant, relative depth estimators can be trained using heterogeneous depth data from various sources [40].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Versatile Depth Estimator Based on Common Relative Depth Estimation and Camera-Specific Relative-to-Metric Depth Conversion

Jun¹,

Lee²

2023

Preprint

View full text Add to dashboard Cite

A typical monocular depth estimator is trained for a single camera, so its performance drops severely on images taken with different cameras. To address this issue, we propose a versatile depth estimator (VDE), composed of a common relative depth estimator (CRDE) and multiple relativeto-metric converters (R2MCs). The CRDE extracts relative depth information, and each R2MC converts the relative information to predict metric depths for a specific camera. The proposed VDE can cope with diverse scenes, including both indoor and outdoor scenes, with only a 1.12% parameter increase per camera. Experimental results demonstrate that VDE supports multiple cameras effectively and efficiently and also achieves state-of-the-art performance in the conventional single-camera scenario.

show abstract

Section: Relative Depth Estimationmentioning

confidence: 99%

Section: Relative-to-metric Depth Conversionmentioning

confidence: 99%

“…However, [35,40] require a large depth-labeled dataset for fitting. Also, [23,39] fine-tune a pre-trained network to estimate metric depth maps for a single camera only.…”

Section: Relative-to-metric Depth Conversionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Versatile Depth Estimator Based on Common Relative Depth Estimation and Camera-Specific Relative-to-Metric Depth Conversion

Jun¹,

Lee²

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…As a result, modern general-purpose SVDE methods trained on stereo data output predictions that cannot be used to recover 3D geometry [17,26,35]. Hence, we describe these methods as not geometry-preserving and their predictions as geometrically incorrect.…”

Section: Introductionmentioning

confidence: 99%

Single-Stage 3D Geometry-Preserving Depth Estimation Model Training on Dataset Mixtures with Uncalibrated Stereo Data

Patakin

Vorontsova

Artemyev

et al. 2022

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Nowadays, robotics, AR, and 3D modeling applications attract considerable attention to single-view depth estimation (SVDE) as it allows estimating scene geometry from a single RGB image. Recent works have demonstrated that the accuracy of an SVDE method hugely depends on the diversity and volume of the training data. However, RGB-D datasets obtained via depth capturing or 3D reconstruction are typically small, synthetic datasets are not photorealistic enough, and all these datasets lack diversity. The large-scale and diverse data can be sourced from stereo images or stereo videos from the web. Typically being uncalibrated, stereo data provides disparities up to unknown shift (geometrically incomplete data), so stereotrained SVDE methods cannot recover 3D geometry. It was recently shown that the distorted point clouds obtained with a stereo-trained SVDE method can be corrected with additional point cloud modules (PCM) separately trained on the geometrically complete data. On the contrary, we propose GP 2 , General-Purpose and Geometry-Preserving training scheme, and show that conventional SVDE models can learn correct shifts themselves without any postprocessing, benefiting from using stereo data even in the geometry-preserving setting. Through experiments on different dataset mixtures, we prove that GP 2 -trained models outperform methods relying on PCM in both accuracy and speed, and report the state-of-the-art results in the general-purpose geometry-preserving SVDE. Moreover, we show that SVDE models can learn to predict geometrically correct depth even when geometrically complete data comprises the minor part of the training set.

show abstract

Relationship Spatialization for Depth Estimation

Qiu

Wang

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Monocular Depth Estimation via Listwise Ranking using the Plackett-Luce Model

Cited by 12 publications

References 19 publications

Versatile Depth Estimator Based on Common Relative Depth Estimation and Camera-Specific Relative-to-Metric Depth Conversion

Versatile Depth Estimator Based on Common Relative Depth Estimation and Camera-Specific Relative-to-Metric Depth Conversion

Single-Stage 3D Geometry-Preserving Depth Estimation Model Training on Dataset Mixtures with Uncalibrated Stereo Data

Relationship Spatialization for Depth Estimation

Contact Info

Product

Resources

About