Filters








985 Hits in 7.0 sec

Comparison of Keypoint Detection Methods for Indoor and Outdoor Scene Recognition

Urszula Libal, Łukasz Łoziuk
2018 Journal of Automation, Mobile Robotics & Intelligent Systems  
This paper presents a simple method for dis nguishing between indoor and outdoor environments in a video sequence. The proposed method is based on the central loca on of keypoints in video frames.  ...  It was observed that the probability distribu ons of selected keypoints are dras cally different between indoor and outdoor environments for all algorithms analyzed.  ...  Classifica on Problem for Scene Recognion Scene recognition can be de ined as a two-class classi ication problem, with class 1 (indoor) and class 2 (outdoor).  ... 
doi:10.14313/jamris_4-2017/32 fatcat:k4rlnyiemvctvdkpskc4aec2xy

Indoor versus outdoor scene recognition for navigation of a micro aerial vehicle using spatial color gist wavelet descriptors

Anitha Ganesan, Anbarasu Balasubramanian
2019 Visual Computing for Industry, Biomedicine, and Art  
Institute of Technology Madras Scene Classification Image Database two, an Indoor-Outdoor Dataset, and the Massachusetts Institute of Technology indoor scene classification dataset [(MIT)-67].  ...  kernel (RBF) kernel and 95.29% and 99.45% using linear kernel for the IITM SCID2 and Indoor-Outdoor datasets, respectively.  ...  Availability of data and materials The datasets used and/or analyzed for indoor versus outdoor scene categorization and indoor scene categorization are available from the corresponding author on reasonable  ... 
doi:10.1186/s42492-019-0030-9 pmid:32240408 fatcat:5ik3pd24izf7toiio343kcu72m

Environment Recognition based on Images using Bag-of-Words

Taurius Petraitis, Rytis Maskeliūnas, Robertas Damaševičius, Dawid Połap, Marcin Woźniak, Marcin Gabryel
2017 Proceedings of the 9th International Joint Conference on Computational Intelligence  
In this paper, scene recognition methods and applications are analysed, and the Bag of Words (BoW), a local image feature based scene classification model is implemented.  ...  This shows that indoor scene classification is a much harder task and a model which does not take into account any mid-level scene information like objects of the scene is not sufficient for the task.  ...  in this category are both outdoor and indoor scenes.  ... 
doi:10.5220/0006585601660176 dblp:conf/ijcci/PetraitisMDPWG17 fatcat:jxctngw4qzfztahz4bq4f3rqiu

Augmented reality application assessment for disseminating rock art

Silvia Blanco-Pons, Berta Carrión-Ruiz, José Luis Lerma
2018 Multimedia tools and applications  
Therefore, the development of AR apps for complex outdoor environments such as rock art sites should be performed with ARToolKit.  ...  However, this method cannot be applied in some complex and outdoor settings such as prehistoric rock art sites owing to the fact that the usage of markers is restricted on site.  ...  Similarly, the authors want to express their gratitude to the General Directorate of Culture and Heritage, Conselleria d'Educació, Investigació, Cultura i Esport, Generalitat Valenciana for letting us  ... 
doi:10.1007/s11042-018-6609-x fatcat:abhw3kvrcjef7fi2mowzccpnpe

Combining Textural and Geometrical Descriptors for Scene Recognition [chapter]

Neslihan Bayramog̃lu, Janne Heikkilä, Matti Pietikäinen
2012 Lecture Notes in Computer Science  
In this paper, we propose combining textural and geometrical descriptors for scene recognition of RGB-D data.  ...  This study represents and evaluates different ways of combining multi-modal descriptors within the BoW approach in the context of indoor scene localization.  ...  Related Work In literature, indoor scene matching is considered to be more challenging than the outdoor scene matching problem [4, 5] because outdoor scenes contain more discriminative and unique features  ... 
doi:10.1007/978-3-642-33868-7_4 fatcat:7o4ahl3peffszho45du5ifgh5y

Voxel Grid-Based Fast Registration of Terrestrial Point Cloud

Biao Xiong, Weize Jiang, Dengke Li, Man Qi
2021 Remote Sensing  
Using different indoor and outdoor data to compare our proposed approach with other 4-point congruent set methods, according to the experimental results, in terms of registration efficiency, the proposed  ...  Obtaining the congruent set of tuples in a large point cloud scene can be challenging.  ...  The results of the voxel grid size of 0.2 m were adopted for indoor data and 0.5 m for outdoor data. This is because the outdoor data scene is large, with a high number of points and plenty of noise.  ... 
doi:10.3390/rs13101905 fatcat:tudgmhky4farnosbdkielwg3u4

Home Interior Classification using SIFT Keypoint Histograms

Brian Ayers, Matthew Boutell
2007 2007 IEEE Conference on Computer Vision and Pattern Recognition  
Semantic scene classification, the process of categorizing photographs into a discrete set of classes using pattern recognition techniques, is a useful ability for image annotation, organization and retrieval  ...  The literature has focused on classifying outdoor scenes such as beaches and sunsets.  ...  outdoor scenes than indoor ones.  ... 
doi:10.1109/cvpr.2007.383485 dblp:conf/cvpr/AyersB07 fatcat:gcbv3524cjda5krs6e5xoctouu

A Synthesis-Based Approach for Thermal-to-Visible Face Verification [article]

Neehar Peri, Joshua Gleason, Carlos D. Castillo, Thirimachos Bourlai, Vishal M. Patel, Rama Chellappa
2021 arXiv   pre-print
Due to the increased cost and difficulty of obtaining diverse, paired thermal and visible spectrum datasets, not many algorithms and large-scale benchmarks for low-light recognition are available.  ...  To the best of our knowledge, with face data from 400 subjects, this dataset represents the most extensive collection of publicly available indoor and long-range outdoor thermal-visible face imagery.  ...  We have implemented a inference engine that uses multiple crops of the detected face and applies a fast implementation of a RANSAC-based method to aggregate the locations of the keypoints at inference  ... 
arXiv:2108.09558v2 fatcat:ve6rtntsxrdexekqryq2wxwy5m

Keypoint induced distance profiles for visual recognition

Tat-Jun Chin, D. Suter
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
This supports supports a coarse-to-fine search strategy for approaches reliant on keypoint matching. We test the idea on 3 datasets for particular scene recognition and report the obtained results.  ...  The distances are computed between properly selected reference keypoints and the keypoints detected in the input image.  ...  Introduction Methods for detecting local interest points or keypoints in images have become reliable workhorses in many scene and object recognition tasks.  ... 
doi:10.1109/cvprw.2009.5206734 fatcat:l7tqku25onfg5f76mbqh7xq3bm

Keypoint induced distance profiles for visual recognition

Tat-Jun Chin, David Suter
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
This supports supports a coarse-to-fine search strategy for approaches reliant on keypoint matching. We test the idea on 3 datasets for particular scene recognition and report the obtained results.  ...  The distances are computed between properly selected reference keypoints and the keypoints detected in the input image.  ...  Introduction Methods for detecting local interest points or keypoints in images have become reliable workhorses in many scene and object recognition tasks.  ... 
doi:10.1109/cvpr.2009.5206734 dblp:conf/cvpr/ChinS09 fatcat:st3ryn3p7fba3k4cj4ygp625di

Early Bird: Loop Closures from Opposing Viewpoints for Perceptually-Aliased Indoor Environments [article]

Satyajit Tourani, Dhagash Desai, Udit Singh Parihar, Sourav Garg, Ravi Kiran Sarvadevabhatla, Michael Milford, K. Madhava Krishna
2020 arXiv   pre-print
Significant advances have been made recently in Visual Place Recognition (VPR), feature correspondence, and localization due to the proliferation of deep-learning-based methods.  ...  For the first time, we demonstrate a localization system capable of state-of-the-art performance despite perceptual aliasing and extreme 180-degree-rotated viewpoint change in a range of real-world and  ...  Similarly, the choice of semantic classes [13] for outdoor environments is not directly transferable to indoor scenes. D.  ... 
arXiv:2010.01421v3 fatcat:3l6f4a3dsnac5jadn3xotwvzdm

RaP-Net: A Region-wise and Point-wise Weighting Network to Extract Robust Features for Indoor Localization [article]

Dongjiang Li, Jinyu Miao, Xuesong Shi, Yuxin Tian, Qiwei Long, Tianyu Cai, Ping Guo, Hongfei Yu, Wei Yang, Haosong Yue, Qi Wei, Fei Qiao
2021 arXiv   pre-print
Various appearance changes between images of the same location are included and can help the model to learn the invariability in typical indoor scenes.  ...  To address the problem, we propose a novel network, RaP-Net, to simultaneously predict region-wise invariability and point-wise reliability, and then extract features by considering both of them.  ...  For example, SURF [20] uses Haar filters and integral images for fast keypoint detection and descriptor calculation.  ... 
arXiv:2012.00234v3 fatcat:vm5bqmi3x5d55j2putisfjfdze

Real-time monocular image-based 6-DoF localization

Hyon Lim, Sudipta N. Sinha, Michael F. Cohen, Matt Uyttendaele, H. Jin Kim
2015 The international journal of robotics research  
We have evaluated our method using ground truth and present results on several challenging indoor and outdoor sequences.  ...  In this paper we present a new real-time image-based localization method for scenes that have been reconstructed offline using structure from motion.  ...  Acknowledgements Part of this work was undertaken while Hyon Lim was visiting Microsoft Research as a research intern.  ... 
doi:10.1177/0278364914561101 fatcat:sfo2sekbtrcvlmw7ktngl2snkq

SuperGlue: Learning Feature Matching With Graph Neural Networks

Paul-Edouard Sarlin, Daniel DeTone, Tomasz Malisiewicz, Andrew Rabinovich
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
SuperGlue outperforms other learned approaches and achieves state-of-the-art results on the task of pose estimation in challenging real-world indoor and outdoor environments.  ...  The proposed method performs matching in real-time on a modern GPU and can be readily integrated into modern SfM or SLAM systems.  ...  Figure 5 : 5 Figure 5: Indoor and outdoor pose estimation.  ... 
doi:10.1109/cvpr42600.2020.00499 dblp:conf/cvpr/SarlinDMR20 fatcat:ltrmgcy3pvdbxgu3nbp4iwdsai

D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features

Xuyang Bai, Zixin Luo, Lei Zhou, Hongbo Fu, Long Quan, Chiew-Lan Tai
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Finally, our method achieves state-ofthe-art results in both indoor and outdoor scenarios, evaluated on 3DMatch and KITTI datasets, and shows its strong generalization ability on the ETH dataset.  ...  In this paper, we leverage a 3D fully convolutional network for 3D point clouds, and propose a novel and practical learning mechanism that densely predicts both a detection score and a description feature  ...  This work is supported by Hong Kong RGC GRF 16206819, 16203518 and Centre for Applied Computing and Interactive Media (ACIM) of School of Creative Media, City University of Hong Kong.  ... 
doi:10.1109/cvpr42600.2020.00639 dblp:conf/cvpr/BaiLZFQT20 fatcat:hxhawmqpirfxlgzegkl2ovmvvq
« Previous Showing results 1 — 15 out of 985 results