PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization

By Alex Kendall, Matthew Koichi Grimes, Roberto Cipolla. 2015

2,475 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, 3D Vision & Pose

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

In 3D Vision & Pose

Citations in the corpus

CPlaNet: Enhancing Image Geolocalization by Combinatorial Partitioning of Maps (2018, methodology): “PoseNet [25, 26] treats pose estimation as a regression problem based on a convolutional neural network.”
PlaNet - Photo Geolocation with Convolutional Neural Networks (2016, background): “PoseNet [28] is a CNN that regresses from a query image to its 6-dof pose.”
A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation (2016, neutral): “). Even for the task like camera re-localization with extremely serious demand for place recognition precision, convolutional features also showed huge improvement compared with SIFT-based localizes (Kendall et al., 2015). Application in place recognition for mobile robots were naturally motivated (”
Attend, Infer, Repeat: Fast Scene Understanding with Generative Models (2016, background): “, object classes [10], camera positions [8] and actions [20].”
Unsupervised Learning of Depth and Ego-Motion from Video (2017, background): “, feature matching [18], pose estimation [26], and stereo [10, 27, 53].”
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016, neutral): “Likewise, it is possible to localize the 6DoF of a camera with regression forest [247] and with deep convolutional neural network [129], and to estimate the depth of a scene (in effect, the map) from a single view solely as a function of the”
From Coarse to Fine: Robust Hierarchical Localization at Large Scale (2018, background): “Another group of approaches attempts to directly regress the pose from the image [20, 18] or to classify it to a spatial bin [46].”
Egocentric Visitors Localization in Cultural Sites (2019, background): “ree common levels of localization are scene recognition [16, 30], location recognition [2, 8, 24, 26] and 6-DOF camera pose estimation [14, 19, 21].”
Fusion and binarization of CNN features for robust topological localization across seasons (2016, background): “small-scale environments [28], in contrast to works where images are matched under substantial appearance changes exploiting GPS metric priors [29].”
Soft Rasterizer: A Differentiable Renderer for Image-Based 3D Reasoning (2019, methodology): “For 3D rigid pose estimation, while early approaches attempt to cast it as classification problem [44], recent approaches [21, 48] can directly regress the 6D pose by using deep neural networks.”
L3-Net: Towards Learning Based LiDAR Localization for Autonomous Driving (2019, background): “PoseNet [15] and its variants [14, 25, 31] attempt to solve the visual re-localization problem, in which an accurate solution is not the goal.”
SuperDepth: Self-Supervised, Super-Resolved Monocular Depth Estimation (2018, methodology): “Following [35], [36], [37], we use the logarithm of a unit quaternion to parameterize the rotation in R(3) and do not require an added normalization constraint unlike previous works [38].”
Egocentric Shopping Cart Localization (2018, methodology): “The standard procedure for the collection of a dataset suitable for imagebased localization involves the acquisition of large sets of images of the environment (this is generally achieved acquiring continuous videos) and the construction of a 3D model using Structure-from-Motion (SfM) algorithms [1]”
Global Pose Estimation with an Attention-Based Recurrent Network (2018, methodology): “PoseNet [22] formulates 6-DoF pose estimation as a regression problem.”