English
Related papers

Related papers: Scene Coordinate Regression with Angle-Based Repro…

200 papers

The goal of this work is to replace objects in an RGB-D scene with corresponding 3D models from a library. We approach this problem by first detecting and segmenting object instances in the scene using the approach from Gupta et al. [13].…

Computer Vision and Pattern Recognition · Computer Science 2015-02-17 Saurabh Gupta , Pablo Arbeláez , Ross Girshick , Jitendra Malik

We apply convolutional neural networks (CNN) to the problem of image orientation detection in the context of determining the correct orientation (from 0, 90, 180, and 270 degrees) of a consumer photo. The problem is especially important for…

Computer Vision and Pattern Recognition · Computer Science 2023-06-02 Ujash Joshi , Michael Guerzhoy

This work presents a novel Convolutional Neural Network (CNN) architecture and a training procedure to enable robust and accurate pose estimation of a noncooperative spacecraft. First, a new CNN architecture is introduced that has scored a…

Computer Vision and Pattern Recognition · Computer Science 2019-09-04 Tae Ha Park , Sumant Sharma , Simone D'Amico

This paper addresses the problem of 3D human pose estimation from single images. While for a long time human skeletons were parameterized and fitted to the observation by satisfying a reprojection error, nowadays researchers directly use…

Computer Vision and Pattern Recognition · Computer Science 2019-03-13 Bastian Wandt , Bodo Rosenhahn

In this paper, we propose a novel approach that learns to sequentially attend to different Convolutional Neural Networks (CNN) layers (i.e., ``what'' feature abstraction to attend to) and different spatial locations of the selected feature…

Computer Vision and Pattern Recognition · Computer Science 2019-06-03 Tony Joseph , Konstantinos G. Derpanis , Faisal Z. Qureshi

Many applications require a camera to be relocalised online, without expensive offline training on the target scene. Whilst both keyframe and sparse keypoint matching methods can be used online, the former often fail away from the training…

Computer Vision and Pattern Recognition · Computer Science 2019-06-21 Tommaso Cavallari , Luca Bertinetto , Jishnu Mukhoti , Philip Torr , Stuart Golodetz

Global visual localization estimates the absolute pose of a camera using a single image, in a previously mapped area. Obtaining the pose from a single image enables many robotics and augmented/virtual reality applications. Inspired by…

Computer Vision and Pattern Recognition · Computer Science 2023-12-19 Mohammad Altillawi , Shile Li , Sai Manoj Prakhya , Ziyuan Liu , Joan Serrat

Many machine learning methods operate by inverting a neural network at inference time, which has become a popular technique for solving inverse problems in computer vision, robotics, and graphics. However, these methods often involve…

Computer Vision and Pattern Recognition · Computer Science 2022-06-22 Ruoshi Liu , Chengzhi Mao , Purva Tendulkar , Hao Wang , Carl Vondrick

Thesedays, Convolutional Neural Networks are widely used in semantic segmentation. However, since CNN-based segmentation networks produce low-resolution outputs with rich semantic information, it is inevitable that spatial details (e.g.,…

Computer Vision and Pattern Recognition · Computer Science 2019-10-03 Youngeun Kim , Seunghyeon Kim , Taekyung Kim , Changick Kim

This paper presents a study on the use of Convolutional Neural Networks for camera relocalisation and its application to map compression. We follow state of the art visual relocalisation results and evaluate the response to different data…

Computer Vision and Pattern Recognition · Computer Science 2018-05-17 Luis Contreras , Walterio Mayol-Cuevas

Training networks to perform metric relocalization traditionally requires accurate image correspondences. In practice, these are obtained by restricting domain coverage, employing additional sensors, or capturing large multi-view datasets.…

Computer Vision and Pattern Recognition · Computer Science 2020-11-03 Mike Kasper , Fernando Nobre , Christoffer Heckman , Nima Keivan

While 360{\deg} cameras offer tremendous new possibilities in vision, graphics, and augmented reality, the spherical images they produce make core feature extraction non-trivial. Convolutional neural networks (CNNs) trained on images from…

Computer Vision and Pattern Recognition · Computer Science 2018-12-10 Yu-Chuan Su , Kristen Grauman

Recently, CNN is a popular choice to handle the hyperspectral image classification challenges. In spite of having such large spectral information in Hyper-Spectral Image(s) (HSI), it creates a curse of dimensionality. Also, large spatial…

Computer Vision and Pattern Recognition · Computer Science 2021-05-06 Jayasree Saha , Yuvraj Khanna , Jayanta Mukherjee

Deep learning has been applied to camera relocalization, in particular, PoseNet and its extended work are the convolutional neural networks which regress the camera pose from a single image. However there are many problems, one of them is…

Robotics · Computer Science 2018-02-27 Qiang Fang , Tianjiang Hu

Scene coordinate regression (SCR) models have proven to be powerful implicit scene representations for 3D vision, enabling visual relocalization and structure-from-motion. SCR models are trained specifically for one scene. If training…

Computer Vision and Pattern Recognition · Computer Science 2025-10-15 Wenjing Bian , Axel Barroso-Laguna , Tommaso Cavallari , Victor Adrian Prisacariu , Eric Brachmann

Learning-based visual relocalizers exhibit leading pose accuracy, but require hours or days of training. Since training needs to happen on each new scene again, long training times make learning-based relocalization impractical for most…

Computer Vision and Pattern Recognition · Computer Science 2023-05-24 Eric Brachmann , Tommaso Cavallari , Victor Adrian Prisacariu

This work investigates the impact of the loss function on the performance of Neural Networks, in the context of a monocular, RGB-only, image localization task. A common technique used when regressing a camera's pose from an image is to…

Computer Vision and Pattern Recognition · Computer Science 2019-07-30 Isaac Ronald Ward , M. A. Asim K. Jalwana , Mohammed Bennamoun

We address the problem of camera pose estimation in visual localization. Current regression-based methods for pose estimation are trained and evaluated scene-wise. They depend on the coordinate frame of the training dataset and show a low…

Computer Vision and Pattern Recognition · Computer Science 2020-11-24 Boris Chidlovskii , Assem Sadek

Scene coordinates regression (SCR), i.e., predicting 3D coordinates for every pixel of a given image, has recently shown promising potential. However, existing methods remain limited to small scenes memorized during training, and thus…

Computer Vision and Pattern Recognition · Computer Science 2023-12-01 Jerome Revaud , Yohann Cabon , Romain Brégier , JongMin Lee , Philippe Weinzaepfel

In this paper, we present a new approach for improving 3D point and line mapping regression for camera re-localization. Previous methods typically rely on feature matching (FM) with stored descriptors or use a single network to encode both…

Computer Vision and Pattern Recognition · Computer Science 2025-03-03 Bach-Thuan Bui , Huy-Hoang Bui , Yasuyuki Fujii , Dinh-Tuan Tran , Joo-Ho Lee