Self-Supervised-depth

by kalilia.

Contents

• Overview
• Datasets
• 1-Monocular-Video: SfM based monocular depth
  • 2017-2019
  • 2020
  • 2021
  • Indoor
• 2-Mutiview: Multi-view-Stereo
• 3-Light-Field-based
• i-Night time depth
• ii-Semantic aware depth
• Related: depth complementation
• Related: Video-depth
• Related: SLAM-Odometry

0-depth-estimation-overview

Conference	Tittle	code	Author	mark	note
	Single Image Depth Estimation: An Overview		Istanbul Technical University	🙉

*-datasets

Tittle	yaer	mark	note
Vision meets Robotics: The KITTI Dataset	2012		Karlsruhe Institute of Technology
NYUDepth-v2：Indoor Segmentation and Support Inference from RGBD Images	2012		New York University
nuScenes: A multimodal dataset for autonomous driving	2018		nuTonomy: an APTIV company

Mono-SfM

2017

Conference	Tittle	code	Author	mark	note
CVPR2017	Semi-Supervised Deep Learning for Monocular Depth Map Prediction		RWTH Aachen University	🙈
CVPR2017	SfMLearner: Unsupervised Learning of Depth and Ego-Motion from Video	link	UC Berkeley	⭐	link

(Back)

2018

Conference	Tittle	code	Author	mark	note
CVPR2018	DVO: Learning Depth from Monocular Videos using Direct Methods		Carnegie Mellon University	🙈
CVPR2018	GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose	link	SenseTime Research	🙈
ECCV2018	DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistency	link	Virginia Tech	🙈
ECCV2018	Supervising the new with the old: learning SFM from SFM	)	University of Oxford	🙈

(Back)

2019

Conference	Tittle	code	Author	mark	note
2019	Self-Supervised 3D Keypoint Learning for Ego-motion Estimation		Toyota Research Institute (TRI)	🙈
ICRA2019	SuperDepth: Self-Supervised, Super-Resolved Monocular Depth Estimation		Toyota Research Institute (TRI)	🙈
AAAI2019	Depth prediction without the sensors: Leveraging structure for unsupervised learning from monocular videos	link	Harvard University/Google Brain	🙈
ICCV2019	Moving indoor: Unsupervised video depth learning in challenging environments.		Tsinghua University	🙈
ICCV2019	Unsupervised High-Resolution Depth Learning From Videos With Dual Networks		Tsinghua University	🙈
ICCV2019	Self-Supervised Monocular Depth Hints	link	Niantic	🙈
ICCV2019	Monodepth2: Digging into self-supervised monocular depth estimation	link	UCL/niantic	🌟
NIPS2019	SC-SfMLearner: Unsupervised scale-consistent depth and ego-motion learning from monocular video		University of Adelaide, Australia	🙈
CVPR2019	Competitive Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentation	link	Max Planck Institute for Intelligent Systems	🙈
CoRL2019	Robust Semi-Supervised Monocular Depth Estimation with Reprojected Distances		Toyota Research Institute (TRI)	🙈

(Back)

2020

Conference	Tittle	code	Author	mark	note
ECCV2020	DeepSFM: Structure From Motion Via Deep Bundle Adjustment		Fudan University	🙈
ECCV2020	P2Net: Patch-match and Plane-regularization for Unsupervised Indoor Depth Estimation		ShanghaiTech Univsertiy	🙈
ECCV2020	Feature-metric Loss for Self-supervised Learning of Depth and Egomotion	link		🙈
CoRL2020	Unsupervised Monocular Depth Learning in Dynamic Scenes		Google Research	🙈
CoRL2020	Attentional Separation-and-Aggregation Network for Self-supervised Depth-Pose Learning in Dynamic Scenes		Tsinghua University	🙉
3DV2020	Neural Ray Surfaces for Self-Supervised Learning of Depth and Ego-motion		Toyota Research Institute (TRI)
ICLR2020	Semantically-Guided Representation Learning for Self-Supervised Monocular Depth		Toyota Research Institute (TRI)
CVPR2020	On the uncertainty of self-supervised monocular depth estimation	link	University of Bologna, Italy	🙈
CVPR2020	Towards Better Generalization: Joint Depth-Pose Learning without PoseNet	link	Tsinghua University	🙈	link
CVPR2020	3D Packing for Self-Supervised Monocular Depth Estimation		Toyota Research Institute (TRI)	🌟	link
CVPR2020	Self-supervised monocular trained depth estimation using self-attention and discrete disparity volume		University of Adelaide	🙈
2020	SAFENet: Self-Supervised Monocular Depth Estimation with Semantic-Aware Feature Extraction	link	Toyota Research Institute (TRI)	🙈
2020	Self-Supervised Monocular Depth Estimation : Solving the Dynamic Object Problem by Semantic Guidance		Technische Universit¨at Braunschweig, Germany	🙈
IROS2020	Toward Hierarchical Self-Supervised Monocular Absolute Depth Estimation for Autonomous Driving Applications	link	Tongji University	🙈

(Back)

2021

Conference	Tittle	code	Author	mark	note
AAAI2021	HR-Depth : High Resolution Self-Supervised Monocular Depth Estimation	link	Zhejiang University	⭐	link
AAAI2021	Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection Consistency	link	KAIST	⭐	link
CVPR2021	Manydepth:The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth	link	Niantic	🙈
CVPR2021	MonoRec: Semi-Supervised Dense Reconstruction in Dynamic Environments from a Single Moving Camera	link	TUM	🙈
IROS2021	Self-Supervised Scale Recovery for Monocular Depth and Egomotion Estimation	link	University of Toronto	🙈
2021	Self-supervised Depth Estimation Leveraging Global Perception and Geometric Smoothness Using On-board Videos		Hong Kong Polytechnic University	🙈
2021	Self-Supervised Structure-from-Motion through Tightly-Coupled Depth and Egomotion Networks		University of Toronto	🙈
2021	Moving SLAM: Fully Unsupervised Deep Learning in Non-Rigid Scenes		HKUST	🙈
2021	Unsupervised Joint Learning of Depth, Optical Flow, Ego-motion from Video		Tongji University	🙈
2021	Monocular Depth Estimation through Virtual-world Supervision and Real-world SfM Self-Supervision			🙈
2021	Self-Supervised Learning of Depth and Ego- Motion from Video by Alternative Training and Geometric Constraints from 3D to 2D			🙈
	-update-time-09-13-2021-
ICCV2021	Fine-grained Semantics-aware Representation Enhancement for Self-supervised Monocular Depth Estimation		Seoul National University	🙈
ICCV2021	Regularizing Nighttime Weirdness: Efficient Self-supervised Monocular Depth Estimation in the Dark		Nanjing University of Science and Technology	🙈
ICCV2021	Self-supervised Monocular Depth Estimation for All Day Images using Domain Separation		Zhejiang University	🙈
ICCV2021	StructDepth: Leveraging the structural regularities for self-supervised indoor depth estimation		Shanghai Jiao Tong University	🙈
ICCV2021	MonoIndoor: Towards Good Practice of Self-Supervised Monocular Depth Estimation for Indoor Environments		OPPO US Research Center	🙈
Sensors Journal 2021	Unsupervised Monocular Depth Perception: Focusing on Moving Objects		Chinese University of Hong Kong	🙈
2021	R4Dyn: Exploring Radar for Self-Supervised Monocular Depth Estimation of Dynamic Scenes		TUM	⭐
2021	Unsupervised Monocular Depth Estimation in Highly Complex Environments		East China University of Science and Technology	🙈
	-update-time-10-13-2021-
3DV 2021	PLNet: Plane and Line Priors for Unsupervised Indoor Depth Estimation		The Chinese University of Hong Kong	🙈
	-update-time-11-29-2021-
2021 BMVC	X-Distill: Improving Self-Supervised Monocular Depth via Cross-Task Distillation
2021 BMVC	Self-Supervised Monocular Depth Estimation with Internal Feature Fusion
2021 3DV	Attention meets Geometry: Geometry Guided Spatial-Temporal Attention for Consistent Self-Supervised Monocular Depth Estimation
ICCV2021	Excavating the Potential Capacity of Self-Supervised Monocular Depth Estimation		Peking University	🙈
2021	SUB-Depth: Self-distillation and Uncertainty Boosting Self-supervised Monocular Depth Estimation

indoor

Conference	Tittle	code	Author	mark	note
	-update-time-01-19-2022-
ICCV2019	Moving Indoor: Unsupervised Video Depth Learning in Challenging Environments		Tsinghua University
ECCV2020	P2Net: Patch-match and Plane-regularizationfor Unsupervised Indoor Depth Estimation		ShanghaiTech Univsertiy
ICCV2021	StructDepth: Leveraging the structural regularities for self-supervised indoordepth estimation
ICCV2021	MonoIndoor: Towards Good Practice of Self-Supervised Monocular Depth Estimation for Indoor Environments		OPPO US Research Cente
3DV2021	PLNet: Plane and Line Priors for Unsupervised Indoor Depth Estimation		The Chinese University of Hong Kong
TIPAMI	Auto-Rectify Network for Unsupervised IndoorDepth Estimation		University of Adelaide
2022	Toward Practical Self-Supervised Monocular Indoor Depth Estimation		University of Southern California
(Back)

3-Multi-view-stereo

Conference	Tittle	code	Author	mark
PAMI2008	SGM：Stereo processing by Semi-Global matching and Mutual Information		German Aerospace Cente	🙈
ECCV2016	Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue		University of Adelaide	🙈
CVPR2017	DispNet: Unsupervised Monocular Depth Estimation with Left-Right Consistency		University College London	🙈
	Cost Volume Pyramid Based Depth Inference for Multi-View Stereo Jiayu	link	Northwestern Polytechnical University	🙈
CVPR2020	Semi-Supervised Deep Learning for Monocular Depth Map Prediction		Australian National University	🙈
AAAI2021	Self-supervised Multi-view Stereo via Effective Co-Segmentation and Data-Augmentation		South China University of Technology	🙈
CVPR2021	Differentiable Diffusion for Dense Depth Estimation from Multi-view Images		Brown University	🙈
ICCV2021	NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereo		Australian National University	⭐

(Back)

4-SLAM-Visual-Odometry

Conference	Tittle	code	Author	mark
ECCV2014	LSD-SLAM: Large-Scale Direct Monocular SLAM		TUM	🙈
TR2015	ORB-SLAM: A Versatile and Accurate Monocular SLAM System		Universidad de Zaragoza	🙈
2016	Direct Visual Odometry using Bit-Planes		Carnegie Mellon University	🙈
TR2017	ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras		Universidad de Zaragoza	🙈
2016	A Photometrically Calibrated Benchmark For Monocular Visual Odometry		TUM	🙈

(Back)

2018

Conference	Tittle	code	Author	mark
PAMI2018	DSO: Direct Sparse Odometry		TUM	🙈
IROS2018	LDSO: Direct Sparse Odometry with Loop Closure		TUM	🙈
ECCV2018	Deep Virtual Stereo Odometry:Leveraging Deep Depth Prediction for Monocular Direct Sparse Odometry		TUM	🙈
2018	Self-improving visual odometry		Magic Leap, Inc.	🙈

(Back)

2019

Conference	Tittle	code	Author	mark
ICLR2019	BA-NET: DENSE BUNDLE ADJUSTMENT NETWORKS		Simon Fraser University	🙈
	TartanVO: A Generalizable Learning-based VO	link	Carnegie Mellon University	🙈
IROS	D2VO: Monocular Deep Direct Visual Odometry			🙈

(Back)

2020

Conference	Tittle	code	Author	mark
ECCV2020	Pseudo RGB-D for Self-Improving Monocular SLAM and Depth Prediction		IIIT-Delhi	🙈
CVPR2020	VOLDOR: Visual Odometry from Log-logistic Dense Optical flow Residuals		Stevens Institute of Technology	🙈
2021	Generalizing to the Open World: Deep Visual Odometry with Online Adaptation		Peking University	🙈
ICRA2021	SA-LOAM: Semantic-aided LiDAR SLAM with Loop Closure		Zhejiang University	🙈

(Back)

Semantic-aware-depth

Conference	Tittle	code	Author	mark
2020	SAFENet: Self-Supervised Monocular Depth Estimation with Semantic-Aware Feature Extraction		KAIST
AAAI2021	Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection Consistency		KAIST
ICCV2021	Fine-grained Semantics-aware Representation Enhancement for Self-supervised Monocular Depth Estimation		Seoul National University

1-Monocular-depth with Cost Volume

Conference	Tittle	code	Author	mark	note
NIPS2020	Forget About the LiDAR: Self-Supervised Depth Estimators with MED Probability Volumes		Korea Advanced Institute of Science and Technology	🙉	link
CVPR2021	DRO: Deep Recurrent Optimizer for Structure-from-Motion		Alibaba A.I. Labs	🙈	link
CVPR2021	The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth	link	Niantic	🙈
CVPR2020	Self-supervised Monocular Trained Depth Estimation using Self-attention and Discrete Disparity Volume	link	Australian Institute for Machine Learning	🙈

(Back)

Video-depth

Conference	Tittle	code	Author	mark
	-update-time-10-13-2021-
CVPR 2020	Consistent Video Depth Estimation		University of Washington
CVPR 2021	Robust Consistent Video Depth Estimation		Facebook
SIGGRAPH 2021	Consistent Depth of Moving Objects in Video

(Back)

Virtual2Real-depth

Conference	Tittle	code	Author	mark

(Back)

NightTime-depth

Conference	Tittle	code	Author	mark
ICCV 2021	Regularizing Nighttime Weirdness: Efficient Self-supervised Monocular Depth Estimation in the Dark		Nanjing University of Science and Technology	🙈
ICCV 2021	Self-supervised Monocular Depth Estimation for All Day Images using Domain Separation		Zhejiang University	🙈
2021	Unsupervised Depth and Ego-motion Estimation for Monocular Thermal Video using Multi-spectral Consistency Loss		KAIST	🙈

(Back)

Light-Filed-based-depth

Conference	Tittle	code	Author	mark
CVPR2021	Differentiable Diffusion for Dense Depth Estimation from Multi-view Images		Brown University	🙈
IROS2021	Unsupervised Learning of Depth Estimation and Visual Odometry for Sparse Light Field Cameras		The University of Sydney	🙈
2021	Occlusion-aware Unsupervised Learning of Depth from 4-D Light Fields		University of Sydney	🙈

(Back)

6-depth-estimation-and-complementation

Conference	Tittle	code	Author	mark
	Sparse Auxiliary Networks for Unified Monocular Depth Prediction and Completion Vitor		Toyota Research Institute (TRI)	🙈
3DV2019	Enhancing self-supervised monocular depth estimation with traditional visual odometry		Univrses AB	🙈
ECCV2020	S3Net: Semantic-aware self-supervised depth estimation with monocular videos and synthetic data		UCSD	🙈
ICCV2021	Unsupervised Depth Completion with Calibrated Backprojection Layers Alex		UCLA	🙈
	-update-time-11-29-2021-
	Self-Supervised Depth Completion for Active Stereo		UCLA	🙈
(Back)