Towards Linear-Time Incremental Structure from Motion

By Changchang Wu. 2013

1,400 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Patches, Planes and Probabilities: A Non-Local Prior for Volumetric 3D Reconstruction (2016, methodology): “As input we assume a set of images and camera poses which we obtain using structure-from-motion [51, 52].”
Reconstructing the World* in Six Days *(As Captured by the Yahoo 100 Million Image Dataset) (2015, background): “Recently, Wu [30] proposed a framework for linear-time SfM.”
Tanks and temples (2017, methodology): “VisualSFM [Wu 2011] is a highly optimized incremental SfM pipeline that integrates ideas from multiple projects [Wu 2013; Wu et al. 2011].”
ACNe: Attentive Context Normalization for Robust Permutation-Equivariant Learning (2020, background): “Their ground truth camera poses are obtained from Structure-from-Motion with VisualSfM [50], from large image collections of publicly available, challenging photo-tourism data.”
VGGT: Visual Geometry Grounded Transformer (2025, neutral): “The traditional SfM pipeline [2, 24, 48, 62, 68, 92] consists of multiple stages, including image matching, triangulation, and bundle adjustment.”
3D all the way: Semantic segmentation of urban scenes from start to end in 3D (2015, neutral): “3.3).”
Structure-from-Motion Revisited (2016, methodology): “We contrast our contributions to the current state-of-the-art systems Bundler (opensource ) [52] and VisualSFM (closed-source) [62].”
Visual SLAM and Structure from Motion in Dynamic Environments (2018, methodology): “While early work in SfM [157] including the prominent “Visual Odometry” [117] made use of the Harris corner detector [57], most recent work [142, 174] employs robust feature detection techniques such as Scale Invariant Feature Transform (SIFT) [99] or its lightweight variants like Speeded Up Robust ”
LF-Net: Learning Local Features from Images (2018, background): “They are versatile [23, 29, 47] and remain useful in many scenarios.”
LIFT: Learned Invariant Feature Transform (2016, methodology): “We used the collections from Piccadilly Circus in London and the Roman Forum in Rome from [26] to reconstruct the 3D using VisualSFM [27], which relies of SIFT features.”
Geometric Loss Functions for Camera Pose Regression with Deep Learning (2017, methodology): “It was collected using a smart phone and structure from motion was used to generate the pose labels [56].”
PGSR: Planar-Based Gaussian Splatting for Efficient and High-Fidelity Surface Reconstruction (2024, neutral): “This paper focuses exclusively on surface reconstruction under given poses, which can be readily computed using SLAM [25], [26], [27] or SFM [28], [29], [30] methods.”
SceneNet: An annotated model generator for indoor scene understanding (2016, methodology): “RGB reconstruction from images of a SceneNet synthetic bedroom and living-room using VisualSfM [3] and [32]. data and we leave it as future work.”
Learning Where to Classify in Multi-view Semantic Segmentation (2014, methodology): “SIFT features [41] are extracted and matched across the images, and reconstructed along with the cameras by using incremental bundle adjustment [42].”
GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose (2018, background): “Lately wide varieties of methods have been proposed in both global and incremental genres [44, 53].”
Recent developments in large-scale tie-point matching (2016, background): “[45] has proposed a different way to restrict a smaller set of interest points.”