3D Visual Phrases for Landmark Recognition

Slides:

Advertisements

Similar presentations

Distinctive Image Features from Scale-Invariant Keypoints

Advertisements

Zhengyou Zhang Vision Technology Group Microsoft Research

Incorporating Site-Level Knowledge to Extract Structured Data from Web Forums Jiang-Ming Yang, Rui Cai, Yida Wang, Jun Zhu, Lei Zhang, and Wei-Ying Ma.

Complex Networks for Representation and Characterization of Images For CS790g Project Bingdong Li 9/23/2009.

An Interactive-Voting Based Map Matching Algorithm

3D Model Matching with Viewpoint-Invariant Patches(VIP) Reporter ：鄒嘉恆 Date ： 10/06/2009.

Presented by Xinyu Chang

Leveraging Stereopsis for Saliency Analysis

3D Shape Histograms for Similarity Search and Classification in Spatial Databases. Mihael Ankerst,Gabi Kastenmuller, Hans-Peter-Kriegel,Thomas Seidl Univ.

Mapping: Scaling Rotation Translation Warp

Localization in indoor environments by querying omnidirectional visual maps using perspective images Miguel Lourenco, V. Pedro and João P. Barreto ICRA.

Image alignment Image from

One-Shot Multi-Set Non-rigid Feature-Spatial Matching

A Study of Approaches for Object Recognition

Scale Invariant Feature Transform (SIFT)

P. Rodríguez, R. Dosil, X. M. Pardo, V. Leborán Grupo de Visión Artificial Departamento de Electrónica e Computación Universidade de Santiago de Compostela.

Presenting by, Prashanth B R 1AR08CS035 Dept.Of CSE. AIeMS-Bidadi. Sketch4Match – Content-based Image Retrieval System Using Sketches Under the Guidance.

Computer vision: models, learning and inference

Wang, Z., et al. Presented by: Kayla Henneman October 27, 2014 WHO IS HERE: LOCATION AWARE FACE RECOGNITION.

Yuping Lin and Gérard Medioni.  Introduction  Method  Register UAV streams to a global reference image ▪ Consecutive UAV image registration ▪ UAV to.

Computer vision.

Final Exam Review CS485/685 Computer Vision Prof. Bebis.

Assembler Efficient Discovery of Spatial Co-evolving Patterns in Massive Geo-sensory Data Sheng QIAN SIGKDD 2015.

AUTOMATIC ANNOTATION OF GEO-INFORMATION IN PANORAMIC STREET VIEW BY IMAGE RETRIEVAL Ming Chen, Yueting Zhuang, Fei Wu College of Computer Science, Zhejiang.

Recognition and Matching based on local invariant features Cordelia Schmid INRIA, Grenoble David Lowe Univ. of British Columbia.

Characterizing activity in video shots based on salient points Nicolas Moënne-Loccoz Viper group Computer vision & multimedia laboratory University of.

General Tensor Discriminant Analysis and Gabor Features for Gait Recognition by D. Tao, X. Li, and J. Maybank, TPAMI 2007 Presented by Iulian Pruteanu.

MESA LAB Multi-view image stitching Guimei Zhang MESA LAB MESA (Mechatronics, Embedded Systems and Automation) LAB School of Engineering, University of.

Local Non-Negative Matrix Factorization as a Visual Representation Tao Feng, Stan Z. Li, Heung-Yeung Shum, HongJiang Zhang 2002 IEEE Presenter : 張庭豪.

A Two-level Pose Estimation Framework Using Majority Voting of Gabor Wavelets and Bunch Graph Analysis J. Wu, J. M. Pedersen, D. Putthividhya, D. Norgaard,

Evaluation of interest points and descriptors. Introduction Quantitative evaluation of interest point detectors –points / regions at the same relative.

Visual Categorization With Bags of Keypoints Original Authors: G. Csurka, C.R. Dance, L. Fan, J. Willamowski, C. Bray ECCV Workshop on Statistical Learning.

Shiliang Zhang1, Qi Tian2, Gang Hua3, Qingming Huang4, Shipeng Li2 1Key Lab of Intelli. Info. Process., Inst. of Comput. Tech., CAS, Beijing , China.

Large Scale Time-Varying Data Visualization Han-Wei Shen Department of Computer and Information Science The Ohio State University.

Silhouette Segmentation in Multiple Views Wonwoo Lee, Woontack Woo, and Edmond Boyer PAMI, VOL. 33, NO. 7, JULY 2011 Donguk Seo

Line Matching Jonghee Park GIST CV-Lab..  Lines –Fundamental feature in many computer vision fields 3D reconstruction, SLAM, motion estimation –Useful.

Image features and properties. Image content representation The simplest representation of an image pattern is to list image pixels, one after the other.

Face recognition using Histograms of Oriented Gradients

Karel Lebeda, Simon Hadfield, Richard Bowden

Deeply learned face representations are sparse, selective, and robust

Image Retrieval and Annotation via a Stochastic Modeling Approach

JPEG Compressed Image Retrieval via Statistical Features

Source: Pattern Recognition, 37(5), P , 2004

Scale Invariant Feature Transform (SIFT)

Range Image Segmentation for Modeling and Object Detection in Urban Scenes Cecilia Chen & Ioannis Stamos Computer Science Department Graduate Center, Hunter.

Capturing, Processing and Experiencing Indian Monuments

Gender Classification Using Scaled Conjugate Gradient Back Propagation

Learning Mid-Level Features For Recognition

Multi-perspective Panoramas

Paper Presentation: Shape and Matching

Final Year Project Presentation --- Magic Paint Face

Li Fei-Fei, UIUC Rob Fergus, MIT Antonio Torralba, MIT

Object Recognition in the Dynamic Link Architecture

Feature Space Based Watermarking in Multi-Images

Level Set Tree Feature Detection

CAP 5415 Computer Vision Fall 2012 Dr. Mubarak Shah Lecture-5

Shape matching and object recognition using shape contexts

RGB-D Image for Scene Recognition by Jiaqi Guo

Institute of Neural Information Processing (Prof. Heiko Neumann •

Aim of the project Take your image Submit it to the search engine

Brief Review of Recognition + Context

Paper Reading Dalong Du April.08, 2011.

Outline Announcement Perceptual organization, grouping, and segmentation Hough transform Read Chapter 17 of the textbook File: week14-m.ppt.

Descriptions of 3-D Objects and Scenes

Boolean Operations for Free-form Models Represented in Geometry Images

ECE734 Project-Scale Invariant Feature Transform Algorithm

Multi-Information Based GCPs Selection Method

Recognition and Matching based on local invariant features

Authors: J.J. Murillo-Fuentes

Problem Image and Volume Segmentation:

Presentation transcript:

3D Visual Phrases for Landmark Recognition Qiang Hao, Rui Cai, Zhiwei Li, Lei Zhang, Yanwei Pang, Feng Wu Tianjin University, Tianjin 300072, P.R. China Microsoft Research Asia, Beijing 100080, P.R. China

Outline Introduction 3D Visual Phrase Discovery 3D Visual Phrase Description 3D Visual Phrase Detection Evaluation Conclusion and Future Work

Introduction Most existing work(BoW) features extracted from irrelevant objects treats the database images independently A 3D visual phrase is a triangular facet on the surface of a reconstructed 3D landmark model. explicitly characterize the spatial structure of a 3D object highly robust to projective transformations due to viewpoint changes.

Introduction

3D Visual Phrase Discovery 3D Landmark Reconstruction 3D Point Selection 3D Visual Phrase Generation Multi-Scale 3D Visual Phrases

3D Visual Phrase Discovery

Multi-Scale 3D Visual Phrases

3D Visual Phrase Description Visual Appearance Geometric Structure

Visual Appearance

Geometric Structure

3D Visual Phrase Detection Appearance-based Point Matching Geometry-based Intra-Phrase Ranking Graph-based Inter-Phrase Refinement

3D Visual Phrase Detection

Evaluation

Evaluation

Evaluation

Evaluation

Conclusion and Future Work In contrast to 2D visual phrases defined in 2D image planes, 3D visual phrases are derived from the physical space and explicitly characterize the 3D spatial structure of a landmark Highly robust to viewpoint changes. Geometric constraintsare desired to afford to more relaxed point matching Accelerate the algorithms, especially the point matching step.

Thank you for listening