From single image query to detailed 3D reconstruction

By Johannes L. Schönberger, Filip Radenovic, Ondřej Chum, Jan-Michael Frahm. 2015

125 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

Fine-Tuning CNN Image Retrieval with No Human Annotation (2017, methodology): “The state-ofthe-art retrieval-SfM pipeline [17] takes an unordered image collection as input and attempts to build all possible 3Dmodels.”
Working hard to know your neighbor's margins: Local descriptor learning loss (2017, unknown): “image retrieval [1, 2], panorama stitching [3], wide baseline stereo [4], 3D-reconstruction [5, 6].”
Learning and aggregating deep local descriptors for instance-level recognition (2020, background): “It is the outcome of Stucture-from-Motion (SfM) [40] with 551 3D models for training.”
Pixelwise View Selection for Unstructured Multi-View Stereo (2016, unknown): “Large-scale 3D reconstruction from Internet photos has seen a tremendous evolution in sparse modeling using Structure-from-Motion (SfM) [1,10,21,37,39,40,45, 61] and in dense modeling using Multi-View Stereo (MVS) [2,12,13,15,42,43,59].”
CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples (2016, methodology): “zoom in, zoom out [53,54], sideways crawl [40], help to build larger and complete model.”
Structure-from-Motion Revisited (2016, background): “Inspired by these works, increasingly largescale reconstruction systems have been developed for hundreds of thousands [1] and millions [20, 62, 51, 50] to recently a hundred million Internet photos [30].”
Reconstructing the World* in Six Days *(As Captured by the Yahoo 100 Million Image Dataset) (2015, background): “In practice, these goals have been prioritized differently by existing LS-SfM frameworks [25, 26, 4, 3, 9, 24].”
Privacy Preserving Image-Based Localization (2019, background): “Localizing a device within a scene by computing the camera pose from an image is a fundamental problem in computer vision, with high relevance in applications such as robotics [16, 19, 65], augmented/mixed reality (AR/MR) [36, 46], and structure from motion (SfM) [25, 61, 63].”
Training Vision Transformers for Image Retrieval (2021, unknown): “For training, we use the SFM120k dataset (Radenovi´c et al., 2018b) which is obtained by applying structure-from-motion and 3D reconstruction to large unlabelled image collections (Schonberger et al., 2015).”
Comparative Evaluation of Hand-Crafted and Learned Local Features (2017, background): “The descriptor matching uses an efficient GPU implementation, and we use COLMAP [37,40] for the SFM and MVS evaluation, while CMVS [11] is used to cluster the larger datasets into more manageable image clusters for the dense reconstruction.”
MegaDepth: Learning Single-View Depth Prediction from Internet Photos (2018, methodology): “Using such methods, past work has used retrieval and SfM to build a 3D model seeded from a single image [31], or registered a photo to an existing 3D model to transfer depth [40].”
Mining on Manifolds: Metric Learning Without Labels (2018, background): “This is appropriate since the geometry of the scene is known, but the whole process assumes the existence of another computer vision system based on local descriptors, which is rather expensive [46].”