Back
Point clouds
Structure from motion (SFM) definition
Estimates a scene's 3D structure from a collection of 2D photographs.
Structure from motion vs Photogrammetry
Unlike photogrammetry, the structure from motion approach does not require prior knowledge of 3D position, camera orientation, or control point information. A least square bundle adjustment technique aligns photos and produces a sparse point cloud reflecting the most prominent features in the images.
Resources (open source)
Back to Hub
#
code
description / paper / conference
language
framework
framework
license
tags
2
An unsupervised learning framework for depth and ego motion estimation from monocular videos
25/04/2017
this week
this week
Unsupervised Learning of Depth and Ego Motion from Video
We present an unsupervised learning framework for the task of monocular depthand camera motion estimation from unstructured video sequences. We achieve thisby simultaneously training depth and camera pose...
conferenceJupyter
Tensorflow
Tensorflow
neural-networks depth-prediction self-supervised-learning unsupervised-learning visual-odometry
3
Bundler SFM
10/03/2013
this week
this week
Bundler Structure from Motion Toolkit
C
3d-reconstruction 3d-vision bundle-adjustment computer-vision computer-vision-tools
4
TRI ML Monocular Depth Estimation Repository
06/12/2019
this week
this week
Sparse Auxiliary Networks for Unified Monocular Depth Prediction and Completion
homepage
Neural Ray Surfaces for Self Supervised Learning of Depth and Ego motion
Semantically Guided Representation Learning for Self Supervised Monocular Depth
homepage
3D Packing for Self Supervised Monocular Depth Estimation
homepage
Estimating scene geometry from data obtained with cost-effective sensors is key for robots and self-driving cars. In this paper, we study the problem of predicting dense depth from a single RGB image (monodepth)...
conferencehomepage
Neural Ray Surfaces for Self Supervised Learning of Depth and Ego motion
Self-supervised learning has emerged as a powerful tool for depth and ego-motion estimation, leading to state-of-the-art results on benchmark datasets. However, one significant limitation shared by current...
homepageSemantically Guided Representation Learning for Self Supervised Monocular Depth
Self-supervised learning is showing great promise for monocular depth estimation, using geometry as the only source of supervision. Depth networks are indeed capable of learning representations that relate...
conferencehomepage
3D Packing for Self Supervised Monocular Depth Estimation
Although cameras are ubiquitous, robotic platforms typically rely on active sensors like LiDAR for direct 3D perception. In this work, we propose a novel self-supervised monocular depth estimation method...
conferencehomepage
Python
Pytorch
Pytorch
5
Pytorch version of SFMLearner from Tinghui Zhou et al.
19/10/2017
this week
this week
Unsupervised Learning of Depth and Ego Motion from Video
We present an unsupervised learning framework for the task of monocular depthand camera motion estimation from unstructured video sequences. We achieve thisby simultaneously training depth and camera pose...
conferencePython
Pytorch
Pytorch
depth disparity kitti unsupervised
6
SFM Visual SLAM
26/01/2016
this week
this week
Pytorch
augmented-reality awesome-list slam visual-inertial visual-odometry
7
Unsupervised Scale consistent Depth Learning from Video (IJCV2021 & NeurIPS 2019)
30/08/2019
this week
this week
Unsupervised Scale consistent Depth and Ego motion Learning from Monocular Video
homepage
Unsupervised Scale consistent Depth Learning from Video
Recent work has shown that CNN-based depth and ego-motion estimators can be learned using unlabelled monocular videos. However, the performance is limited by unidentified moving objects that violate the...
conferencehomepage
Unsupervised Scale consistent Depth Learning from Video
We propose a monocular depth estimator SC-Depth, which requires only unlabelled videos for training and enables the scale-consistent prediction at inference time. Our contributions include: (i) we propose...
homepagePython
Pytorch
Pytorch
depth-estimation kitti nyuv2-dataset sc-sfmlearner visual-odometry
8
Pixel Perfect Structure from Motion with Featuremetric Refinement (ICCV 2021, Oral)
17/08/2021
this week
this week
Pixel Perfect Structure from Motion with Featuremetric Refinement
Finding local features that are repeatable across multiple views is a cornerstone of sparse 3D reconstruction. The classical image matching paradigm detects keypoints per-image once and for all, which...
conferenceC++
Pytorch
Pytorch
9
Graph Based Parallel Large Scale Structure from Motion
While Structure from Motion (SFM) achieves great success in 3D reconstruction, it still meets challenges on large scale scenes. In this work, large scale SFM is deemed as a graph problem, and we tackle...
Tensorflow
10
C3DPO Canonical 3D Pose Networks for Non rigid Structure From Motion.
11/09/2019
this week
this week
C3DPO: Canonical 3D Pose Networks for Non Rigid Structure From Motion
We propose C3DPO, a method for extracting 3D models of deformable objects from 2D keypoint annotations in unconstrained images. We do so by learning a deep network that reconstructs a 3D object from a...
conferencePython
Pytorch
Pytorch
11
ENFT: Efficient Non Consecutive Feature Tracking for Robust Structure from Motion
Structure-from-motion (SFM) largely relies on feature tracking. In imagesequences, if disjointed tracks caused by objects moving in and out of thefield of view, occasional occlusion, or image noise, are...
12
This is a Pytorch implementation of the ECCV2020 paper "DeepSFM: Structure From Motion Via Deep Bundle Adjustment".
10/08/2020
this week
this week
DeepSFM: Structure From Motion Via Deep Bundle Adjustment
Structure from motion (SFM) is an essential computer vision problem which has not been well handled by deep learning. One of the promising trends is to apply explicit structural constraint, e.g. 3D cost...
conferencePython
Pytorch
Pytorch
13
(ECCV 2020) Stochastic Frequency Masking to Improve Super Resolution and Denoising Networks
16/03/2020
2 weeks ago
2 weeks ago
Stochastic Frequency Masking to Improve Super Resolution and Denoising Networks
Super-resolution and denoising are ill-posed yet fundamental image restoration tasks. In blind settings, the degradation kernel or the noise level are unknown. This makes restoration even more challenging,...
conferencePython
Pytorch
Pytorch
14
Deep SFM Revisited
17/06/2021
this week
this week
Deep Two View Structure from Motion Revisited
Two-view structure-from-motion (SFM) is the cornerstone of 3D reconstruction and visual SLAM. Existing deep learning-based approaches formulate the problem by either recovering absolute pose scales from...
conferencePython
Pytorch
Pytorch
computer-vision neural-networks
15
[ECCV'20] Patch match and Plane regularization for Unsupervised Indoor Depth Estimation
15/07/2020
this week
this week
PNet: Patch match and Plane regularization for Unsupervised Indoor Depth Estimation
P$^{2}$Net: Patch match and Plane regularization for Unsupervised Indoor Depth Estimation
This paper tackles the unsupervised depth estimation task in indoor environments. The task is extremely challenging because of the vast areas of non-texture regions in these scenes. These areas could overwhelm...
P$^{2}$Net: Patch match and Plane regularization for Unsupervised Indoor Depth Estimation
This paper tackles the unsupervised depth estimation task in indoor environments. The task is extremely challenging because of the vast areas of non-texture regions in these scenes. These areas could overwhelm...
Python
Pytorch
Pytorch
depth-estimation eccv2020 extract-superpixel indoor nyuv2 pose-estimation scannet self-supervised unsupervised-learning
16
Ridge SFM Structure from Motion via robust pairwise matching under depth uncertainty
21/10/2020
this week
this week
RidgeSFM: Structure from Motion via Robust Pairwise Matching Under Depth Uncertainty
We consider the problem of simultaneously estimating a dense depth map and camera pose for a large set of images of an indoor scene. While classical SFM pipelines rely on a two-step approach where cameras...
Python
Pytorch
Pytorch
Other
17
LineSFM
20/03/2017
2 weeks ago
2 weeks ago
Robust SFM with Little Image Overlap
Usual Structure-from-Motion (SFM) techniques require at least trifocaloverlaps to calibrate cameras and reconstruct a scene. We consider herescenariOS of reduced image sets with little overlap, possibly...
C++
19
Code for "PackNet SFM: 3D Packing for Self Supervised Monocular Depth Estimation"
30/04/2019
10 months ago
10 months ago
3D Packing for Self Supervised Monocular Depth Estimation
Although cameras are ubiquitous, robotic platforms typically rely on active sensors like LiDAR for direct 3D perception. In this work, we propose a novel self-supervised monocular depth estimation method...
conferencemonocular-depth pose-estimation self-supervised-learning
20
Equivariant SFM
16/08/2021
1 month ago
1 month ago
Deep Permutation Equivariant Structure from Motion
Existing deep methods produce highly accurate 3D reconstructions in stereo and multiview stereo settings, i.e., when cameras are both internally and externally calibrated. Nevertheless, the challenge of...
conferencePython
Pytorch
Pytorch
21
The repository for paper https://arxiv.Org/abs/1902.10840
02/03/2019
1 month ago
1 month ago
Deep Interpretable Non Rigid Structure from Motion
All current non-rigid structure from motion (NRSFM) algorithms are limitedwith respect to: (i) the number of images, and (ii) the type of shapevariability they can handle. This has hampered the practical...
Python
Tensorflow
Tensorflow
23
Adaptive State Frequency Memory Recurrent Neural Network
08/02/2018
2 months ago
2 months ago
State Frequency Memory Recurrent Neural Networks
Modeling temporal sequences plays a fundamental role in various modern applications and has drawn more and more attentions in the machine learning community. Among those efforts on improving the capability...
Python
Tensorflow
Keras
Tensorflow
Keras
recurrent-neural-networks
24
Code for NeurIPS 2019 paper "From voxels to pixels and back: Self supervision in natural image reconstruction from fMRI"
25/10/2019
3 months ago
3 months ago
From voxels to pixels and back: Self supervision in natural image reconstruction from fMRI
Reconstructing observed images from fMRI brain recordings is challenging. Unfortunately, acquiring sufficient "labeled" pairs of {Image, fMRI} (i.e., images with their corresponding fMRI responses) to...
conferencePython
Tensorflow
Tensorflow
Other
25
Code for CVPR 2017 paper Distinguishing the Indistinguishable: Exploring Structural Ambiguities via Geodesic Context.
12/09/2018
1 week ago
1 week ago
Distinguishing the Indistinguishable: Exploring Structural Ambiguities via Geodesic Context
A perennial problem in structure from motion (SFM) is visual ambiguity posed by repetitive structures. Recent disambiguating algorithms infer ambiguities mainly via explicit background context, thus face...
homepageC++
3d-reconstruction bundle-adjustment bundler-sfm cvpr-2017 disambiguation image-based-modeling multiview-stereo multiview-triangulation point-cloud
26
[MedIA 2022] Depth and ego motion estimation in endoscopy
07/10/2021
2 weeks ago
2 weeks ago
Self Supervised Monocular Depth and Ego Motion Estimation in Endoscopy: Appearance Flow to the Rescue
Recently, self-supervised learning technology has been applied to calculate depth and ego-motion from monocular videos, achieving remarkable performance in autonomous driving scenariOS. One widely adopted...
Python
Pytorch
Pytorch
brightness-calibration computer-vision neural-networks depth-estimation endoscopy monodepth pose-estimation self-supervision
27
Code for R MSFM
26/07/2021
1 month ago
1 month ago
R MSFM: Recurrent Multi Scale Feature Modulation for Monocular Depth Estimating
In this paper, we propose Recurrent Multi-Scale Feature Modulation (R-MSFM), a new deep network architecture for self-supervised monocular depth estimation. R-MSFM extracts per-pixel features, builds...
Python
Pytorch
Pytorch
29
SFM twitter harvester
12/11/2015
4 months ago
4 months ago
A harvester for twitter content as part of Social Feed Manager.
Python
social-feed-manager
30
Local Non Rigid Structure From Motion From Diffeomorphic Mappings
We propose a new formulation to non-rigid structure-from-motion that only requires the deforming surface to preserve its differential structure. This is a much weaker assumption than the traditional ones...
31
Extreme Structure From Motion for Indoor Panoramas Without Visual Overlaps
This paper proposes an extreme structure from motion (SFM) algorithm for residential indoor panoramas that have little to no visual overlaps. Only a single panorama is present in a room for many cases,...
Pytorch
32
GPSFM: Global Projective SFM Using Algebraic Constraints on Multi View Fundamental Matrices
This paper addresses the problem of recovering projective camera matrices from collections of fundamental matrices in multiview settings. We make two main contributions. First, given ${n \choose 2}$ fundamental...
conference34
Unsupervised Learning of Depth and Ego Motion from Cylindrical Panoramic Video
Unsupervised Learning of Depth and Ego Motion from Cylindrical Panoramic Video with Applications for Virtual Reality
We introduce a convolutional neural network model for unsupervised learning of depth and ego-motion from cylindrical panoramic video. Panoramic depth estimation is an important technology for applications...
Unsupervised Learning of Depth and Ego Motion from Cylindrical Panoramic Video with Applications for Virtual Reality
We introduce a convolutional neural network model for unsupervised learning of depth and ego-motion from cylindrical panoramic video. Panoramic depth estimation is an important technology for applications...
Tensorflow
35
Deep NRSFM++: Towards Unsupervised 2D 3D Lifting in the Wild
The recovery of 3D shape and pose from 2D landmarks stemming from a large ensemble of images can be viewed as a non-rigid structure from motion (NRSFM) problem. Classical NRSFM approaches, however, are...
36
Unsupervised Depth and Ego motion Estimation for Monocular Thermal Video using Multi spectral Consistency Loss
Most of the deep-learning based depth and ego-motion networks have been designed for visible cameras. However, visible cameras heavily rely on the presence of an external light source. Therefore, it is...
Pytorch
37
PCN: Point Completion Network
ASFM Net: Asymmetrical Siamese Feature Matching Network for Point Completion
Shape completion, the problem of estimating the complete geometry of objects from partial observations, lies at the core of many vision and robotics applications. In this work, we propose Point Completion...
ASFM Net: Asymmetrical Siamese Feature Matching Network for Point Completion
We tackle the problem of object completion from point clouds and propose a novel point cloud completion network employing an Asymmetrical Siamese Feature Matching strategy, termed as ASFM-Net. Specifically,...
Tensorflow
41
Joint Optimization of Hadamard Sensing and Reconstruction in Compressed Sensing Fluorescence Microscopy
Compressed sensing fluorescence microscopy (CS-FM) proposes a scheme whereby less measurements are collected during sensing and reconstruction is performed to recover the image. Much work has gone into...
Pytorch
42
Towards Continual, Online, Unsupervised Depth
Although depth extraction with passive sensors has seen remarkable improvement with deep learning, these approaches may fail to obtain correct depth if they are exposed to environments not observed during...
Pytorch
43
An interpretable machine learning framework for dark matter halo formation
Machine learning cosmological structure formation
Halo and Galaxy Formation Histories from the Millennium Simulation: Public release of a VO oriented and SQL queryable database for studying the evolution of galaxies in the LambdaCDM cosmogony
We present a generalization of our recently proposed machine learning framework, aiming to provide new physical insights into dark matter halo formation. We investigate the impact of the initial density...
Machine learning cosmological structure formation
We train a machine learning algorithm to learn cosmological structureformation from N-body simulations. The algorithm infers the relationshipbetween the initial conditions and the final dark matter haloes,...
Halo and Galaxy Formation Histories from the Millennium Simulation: Public release of a VO oriented and SQL queryable database for studying the evolution of galaxies in the LambdaCDM cosmogony
The Millennium Run is the largest simulation of the formation of structure within the $\Lambda$CDM cosmogony so far carried out. It uses $10^{10}$ particles to follow the dark matter distribution in a...
44
Function on Function Regression for the Identification of Epigenetic Regions Exhibiting Windows of Susceptibility to Environmental Exposures
The ability to identify time periods when individuals are most susceptible to exposures, as well as the biological mechanisms through which these exposures act, is of great public health interest. Growing...
47
Stochastic Frequency Masking to Improve Super Resolution and Denoising Networks
Super-resolution and denoising are ill-posed yet fundamental image restoration tasks. In blind settings, the degradation kernel or the noise level are unknown. This makes restoration even more challenging,...
conferencePytorch
48
Code accompanying Tooley et al. (2020), Associations between Neighborhood SES & Functional Brain Development. https://doi.org/10.1093/cercor/bhz066
49
Observing the host galaxies of high redshift quasars with JWST: predictions from the BlueTides simulation
The bright emission from high-redshift quasars completely conceals their host galaxies in the rest-frame ultraviolet/optical, with detection of the hosts in these wavelengths eluding even the Hubble Space...
50
Fast Decomposable Submodular Function Minimization using Constrained Total Variation
We consider the problem of minimizing the sum of submodular set functions assuming minimization oracles of each summand function. Most existing approaches reformulate the problem as the convex minimization...
conference51
54
VoxelNet: End to End Learning for Point Cloud Based 3D Object Detection
Frustum PointNets for 3D Object Detection from RGB D Data
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
PCPNET: Learning Local Shape Properties from Raw Point Clouds
Accurate detection of objects in 3D point clouds is a central problem in manyapplications, such as autonomous navigation, housekeeping robots, andaugmented/virtual reality. To interface a highly sparse...
conferenceFrustum PointNets for 3D Object Detection from RGB D Data
In this work, we study 3D object detection from RGB-D data in both indoor andoutdoor scenes. While previous methods focus on images or 3D voxels, oftenobscuring natural 3D patterns and invariances of 3D...
conferencePointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
Point cloud is an important type of geometric data structure. Due to itsirregular format, most researchers transform such data to regular 3D voxelgrids or collections of images. This, however, renders...
conferencePCPNET: Learning Local Shape Properties from Raw Point Clouds
In this paper, we propose PCPNet, a deep-learning based approach forestimating local 3D shape properties in point clouds. In contrast to themajority of prior techniques that concentrate on global or mid-levelattributes,...
Tensorflow
56
Unsupervised Odometry and Depth Learning for Endoscopic Capsule Robots
In the last decade, many medical companies and research groups have tried toconvert passive capsule endoscopes as an emerging and minimally invasivediagnostic technology into actively steerable endoscopic...
Tensorflow
57
59
Approximate Decomposable Submodular Function Minimization for Cardinality Based Components
Minimizing a sum of simple submodular functions of limited support is a special case of general submodular function minimization that has seen numerous applications in machine learning. We develop fast...
conference60
Revisiting Decomposable Submodular Function Minimization with Incidence Relations
We introduce a new approach to decomposable submodular function minimization(DSFM) that exploits incidence relations. Incidence relations describe whichvariables effectively influence the component functions,...
conferenceBack to Hub