Current state of supervised and unsupervised depth completion methods

Last update: Dec 28, 2022

Overview

Awesome Depth Completion

Table of Contents

About Sparse-to-Dense Depth Completion
Current State of Depth Completion

About Sparse-to-Dense Depth Completion

In the sparse-to-dense depth completion problem, one wants to infer the dense depth map of a 3-D scene given an RGB image and its corresponding sparse reconstruction in the form of a sparse depth map obtained either from computational methods such as SfM (Strcuture-from-Motion) or active sensors such as lidar or structured light sensors.

Example 1: VOID dataset (indoor VIO)

Input RGB image	Sparse point cloud	Output point cloud from KBNet

Example 2: KITTI dataset (outdoor lidar)

Input RGB image	Output point cloud from ScaffNet

Current State of Depth Completion Methods

Here we compile both unsupervised/self-supervised (monocular and stereo) and supervised methods published in recent conferences and journals on the VOID (Wong et. al., 2020) and KITTI (Uhrig et. al., 2017) depth completion benchmarks. Our ranking considers all four metrics rather than just RMSE.

Quick Links

Unsupervised VOID Benchmark
Supervised VOID Benchmark
Unsupervised KITTI Benchmark
Supervised KITTI Benchmark

Unsupervised VOID Depth Completion Benchmark

Paper	Publication	Code	MAE	RMSE	iMAE	iRMSE
Unsupervised Depth Completion with Calibrated Backprojection Layers	ICCV 2021	PyTorch	39.80	95.86	21.16	49.72
Learning Topology from Synthetic Data for Unsupervised Depth Completion	RA-L & ICRA 2021	Tensorflow	60.68	122.01	35.24	67.34
Unsupervised Depth Completion from Visual Inertial Odometry	RA-L & ICRA 2020	Tensorflow	85.05	169.79	48.92	104.02
Dense depth posterior (ddp) from single image and sparse range	CVPR 2019	Tensorflow	151.86	222.36	74.59	112.36
Self-supervised Sparse-to-Dense: Self- supervised Depth Completion from LiDAR and Monocular Camera	ICRA 2019	PyTorch	178.85	243.84	80.12	107.69

Supervised VOID Depth Completion Benchmark

Paper	Publication	Code	MAE	RMSE	iMAE	iRMSE
Scanline Resolution-Invariant Depth Completion Using a Single Image and Sparse LiDAR Point Cloud	RA-L & IROS 2021	N/A	59.40	181.42	19.37	46.56

Unsupervised KITTI Depth Completion Benchmark

Paper	Publication	Code	MAE	RMSE	iMAE	iRMSE
Unsupervised Depth Completion with Calibrated Backprojection Layers	ICCV 2021	PyTorch	256.76	1069.47	1.02	2.95
Learning Topology from Synthetic Data for Unsupervised Depth Completion	RA-L & ICRA 2021	Tensorflow	280.76	1121.93	1.15	3.30
Project to Adapt: Domain Adaptation for Depth Completion from Noisy and Sparse Sensor Data	ACCV 2020	PyTorch	280.42	1095.26	1.19	3.53
Unsupervised Depth Completion from Visual Inertial Odometry	RA-L & ICRA 2020	Tensorflow	299.41	1169.97	1.20	3.56
A Surface Geometry Model for LiDAR Depth Completion	RA-L & ICRA 2021	Tensorflow	298.3	1239.84	1.21	3.76
Dense depth posterior (ddp) from single image and sparse range	CVPR 2019	Tensorflow	343.46	1263.19	1.32	3.58
DFuseNet: Deep Fusion of RGB and Sparse Depth Information for Image Guided Dense Depth Completion	ITSC 2019	PyTorch	429.93	1206.66	1.79	3.62
In Defense of Classical Image Processing: Fast Depth Completion on the CPU	CRV 2018	Python	302.60	1288.46	1.29	3.78
Self-supervised Sparse-to-Dense: Self- supervised Depth Completion from LiDAR and Monocular Camera	ICRA 2019	PyTorch	350.32	1299.85	1.57	4.07
Semantically Guided Depth Upsampling	GCPR 2016	N/A	605.47	2312.57	2.05	7.38

Supervised KITTI Depth Completion Benchmark

Paper	Publication	Code	MAE	RMSE	iMAE	iRMSE
Non-Local Spatial Propagation Network for Depth Completion	ECCV 2020	PyTorch	199.5	741.68	0.84	1.99
CSPN++: Learning Context and Resource Aware Convolutional Spatial Propagation Networks for Depth Completion	AAAI 2020	N/A	209.28	743.69	0.90	2.07
Dense depth posterior (ddp) from single image and sparse range	CVPR 2019	Tensorflow	203.96	832.94	0.85	2.10
Adaptive context-aware multi-modal network for depth completion	TIP 2021	PyTorch	206.80	732.99	0.90	2.08
PENet: Towards Precise and Efficient Image Guided Depth Completion	ICRA 2021	PyTorch	210.55	730.08	0.94	2.17
FCFR-Net: Feature Fusion based Coarse- to-Fine Residual Learning for Depth Completion	AAAI 2021	N/A	217.15	735.81	0.98	2.20
Learning Guided Convolutional Network for Depth Completion	TIP 2020	PyTorch	218.83	736.24	0.99	2.25
DenseLiDAR: A Real-Time Pseudo Dense Depth Guided Depth Completion Network	ICRA 2021	N/A	214.13	755.41	0.96	2.25
A Multi-Scale Guided Cascade Hourglass Network for Depth Completion	WACV 2020	PyTorch	220.41	762.19	0.98	2.30
Sparse and noisy LiDAR completion with RGB guidance and uncertainty	MVA 2019	PyTorch	215.02	772.87	0.93	2.19
A Multi-Scale Guided Cascade Hourglass Network for Depth Completion	WACV 2020	N/A	220.41	762.19	0.98	2.30
Learning Joint 2D-3D Representations for Depth Completion	ICCV 2019	N/A	221.19	752.88	1.14	2.34
DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene From Sparse LiDAR Data and Single Color Image	CVPR 2019	PyTorch	226.50	758.38	1.15	2.56
Depth Completion from Sparse LiDAR Data with Depth-Normal Constraints	ICCV 2019	N/A	235.17	777.05	1.13	2.42
Scanline Resolution-Invariant Depth Completion Using a Single Image and Sparse LiDAR Point Cloud	RA-L & IROS 2021	N/A	233.34	809.09	1.06	2.57
Confidence propagation through cnns for guided sparse depth regression	PAMI 2019	PyTorch	233.26	829.98	1.03	2.60
Self-supervised Sparse-to-Dense: Self- supervised Depth Completion from LiDAR and Monocular Camera	ICRA 2019	PyTorch	249.95	814.73	1.21	2.80
Uncertainty-Aware CNNs for Depth Completion: Uncertainty from Beginning to End	CVPR 2020	PyTorch	251.77	960.05	1.05	3.37
Sparse and Dense Data with CNNs: Depth Completion and Semantic Segmentation	3DV 2019	N/A	234.81	917.64	0.95	2.17
Depth coefficients for depth completion	CVPR 2019	N/A	252.21	988.38	1.13	2.87
Depth estimation via affinity learned with convolutional spatial propagation network	ECCV 2018	N/A	279.46	1019.64	1.15	2.93
Learning morphological operators for depth completion	ACIVS 2019	N/A	310.49	1045.45	1.57	3.84
Sparsity Invariant CNNs	3DV 2017	Tensorflow	416.14	1419.75	1.29	3.25
Deep Convolutional Compressed Sensing for LiDAR Depth Completion	ACCV 2018	Tensorflow	439.48	1325.37	3.19	59.39

Current state of supervised and unsupervised depth completion methods

Related tags

Overview

Awesome Depth Completion

About Sparse-to-Dense Depth Completion

Current State of Depth Completion Methods

Unsupervised VOID Depth Completion Benchmark

Supervised VOID Depth Completion Benchmark

Unsupervised KITTI Depth Completion Benchmark

Supervised KITTI Depth Completion Benchmark

Owner

3D-Transformer: Molecular Representation with Transformer in 3D Space

Code for KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs

Deep learning for spiking neural networks

Read number plates with https://platerecognizer.com/

Non-stationary GP package written from scratch in PyTorch

《Lerning n Intrinsic Grment Spce for Interctive Authoring of Grment Animtion》

EMNLP 2021 Findings' paper, SCICAP: Generating Captions for Scientific Figures

Implementation of StyleSpace Analysis: Disentangled Controls for StyleGAN Image Generation in PyTorch

Tutorial materials for Part of NSU Intro to Deep Learning with PyTorch.

PyTorch implementation of "ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context" (INTERSPEECH 2020)

Relative Positional Encoding for Transformers with Linear Complexity

Repository for the paper titled: "When is BERT Multilingual? Isolating Crucial Ingredients for Cross-lingual Transfer"

Converts geometry node attributes to built-in attributes

Speech Emotion Recognition with Fusion of Acoustic- and Linguistic-Feature-Based Decisions

MapReader: A computer vision pipeline for the semantic exploration of maps at scale

Some bravo or inspiring research works on the topic of curriculum learning.

A library of extension and helper modules for Python's data analysis and machine learning libraries.

CVPR2021 Workshop - HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization.

WarpRNNT loss ported in Numba CPU/CUDA for Pytorch

DIT is a DTLS MitM proxy implemented in Python 3. It can intercept, manipulate and suppress datagrams between two DTLS endpoints and supports psk-based and certificate-based authentication schemes (RSA + ECC).