11,903 Hits in 4.9 sec

Multi-view manhole detection, recognition, and 3D localisation

Radu Timofte, Luc Van Gool
2011 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops)  
We propose an efficient pipeline that starts from images captured by vans, which are then used to detect, recognise and localise the manholes.  ...  We combine 2D and 3D techniques and demonstrate high performance for large amounts of image data.  ...  We thank GeoAutomation for providing the images and for their support.  ... 
doi:10.1109/iccvw.2011.6130242 dblp:conf/iccvw/TimofteG11 fatcat:otf7pfuwszgn3dcktsw3argmia

Multi-camera localisation: a review

Rahul Raman, Sambit Bakshi, Pankaj K. Sa
2013 International Journal of Machine Intelligence and Sensory Signal Processing  
View overlaps in vision-based algorithms have been optimised and several depth measurement techniques have been implemented to extend the localisation from 2D to 3D space.  ...  LED-based techniques like triangulation and LED triangle have given depth measurement alternatives for 3D localisation whereas epipolar geometry has localised cameras with only image information.  ...  object from two cameras.  ... 
doi:10.1504/ijmissp.2013.052876 fatcat:4qca2cmaorcf7glgtgbxgn6zii

Multi-view shape estimation of transparent containers [article]

Alessio Xompero, Ricardo Sanchez-Matilla, Apostolos Modas, Pascal Frossard, Andrea Cavallaro
2020 arXiv   pre-print
The 3D localisation of an object and the estimation of its properties, such as shape and dimensions, are challenging under varying degrees of transparency and lighting conditions.  ...  Under the assumption of circular symmetry along the vertical axis, we estimate the dimensions of an object with a generative 3D sampling model of sparse circumferences, iterative shape fitting and image  ...  LoDE uses an iterative multi-view 3D-2D shape fitting algorithm of a generative 3D sampling model, verifying the model on the object image masks of two wide-baseline cameras.  ... 
arXiv:1911.12354v2 fatcat:lvw4fzmvqzcw7i4yvrfnkt4jqy

Weakly Supervised Training of Monocular 3D Object Detectors Using Wide Baseline Multi-view Traffic Camera Data [article]

Matthew Howe, Ian Reid, Jamie Mackenzie
2021 arXiv   pre-print
To fine-tune the monocular 3D object detector, our method utilises multiple 2D detections from overlapping, wide-baseline views and a loss that encodes the subjacent geometric consistency.  ...  These datasets contain images from a ground level viewpoint, whereas an ideal view for intersection observation would be elevated higher above the road surface.  ...  Monocular 3D Object Detectors: Monocular 3D object detection aims to infer the location of objects in the scene from a single image.  ... 
arXiv:2110.10966v1 fatcat:angwnguodbcfblpblycu6gorzq

3D Pose from Detections [article]

Cosimo Rubino and Marco Crocco and Alessandro Perina and Vittorio Murino and Alessio Del Bue
2015 arXiv   pre-print
We present a novel method to infer, in closed-form, a general 3D spatial occupancy and orientation of a collection of rigid objects given 2D image detections from a sequence of images.  ...  In particular, starting from 2D ellipses fitted to bounding boxes, this novel multi-view problem can be reformulated as the estimation of a quadric (ellipsoid) in 3D.  ...  This will solve for both the 3D localisation and occupancy of each object starting from the image detections in the different views.  ... 
arXiv:1502.04754v3 fatcat:ykxhfbjv6fhgrogb6hrf2da2tq

Vision based Object Recognition and Localization by a Network Connected Distributed Robotic Systems

2012 ELCVIA Electronic Letters on Computer Vision and Image Analysis  
The objects were localised with an error ranging from 2.8cm to 5.2cm from their actual positions in the test arena which has the dimensions of 150x163cm.  ...  The information from both sources, that is ceiling mounted cameras and team of robots, was used collectively to localise the objects in the environment.  ...  Table 1: Object Localisation Information Object ID Actual Position (x,y) Detected Position (x,y) Error (pixels) Localisation error (cm) in the arena of area (150x150cm) Object 1 (3D) (159,931) (163,914  ... 
doi:10.5565/rev/elcvia.486 fatcat:dkazofeqzbfyjhhdvu5q5l5r74

Multi-view traffic sign detection, recognition, and 3D localisation

Radu Timofte, Karel Zimmermann, Luc Van Gool
2009 2009 Workshop on Applications of Computer Vision (WACV)  
The paper proposes a pipeline for the efficient detection and recognition of traffic signs from such images.  ...  The 2D detections in multiple views are subsequently combined to generate 3D hypotheses.  ...  The authors thank GeoAutomation for providing the images.  ... 
doi:10.1109/wacv.2009.5403121 dblp:conf/wacv/TimofteZG09 fatcat:37ehadoiwjgh5pzuxh7if4o5ce

MOLTR: Multiple Object Localisation, Tracking, and Reconstruction from Monocular RGB Videos [article]

Kejie Li, Hamid Rezatofighi, Ian Reid
2021 arXiv   pre-print
Given a new RGB frame, MOLTR firstly applies a monocular 3D detector to localise objects of interest and extract their shape codes that represent the object shapes in a learned embedding space.  ...  Detections are then merged to existing objects in the map after data association.  ...  C. 3D localisation First, MOLTR detects objects of interest given an RGB image.  ... 
arXiv:2012.05360v2 fatcat:h7yhnpd6ynft7kdqwcuu6xwqcq

VALUE: Large Scale Voting-based Automatic Labelling for Urban Environments [article]

Giacomo Dabisias, Emanuele Ruffaldi, Hugo Grimmett, Peter Ondruska
2020 arXiv   pre-print
This paper presents a simple and robust method for the automatic localisation of static 3D objects in large-scale urban environments.  ...  To evaluate the method we collected city-scale data sets from New York City and San Francisco consisting of almost 400k images spanning the area of 40 km^2 and used it to accurately recover the 3D positions  ...  , a multi-view triangulation approach can be used [11] .  ... 
arXiv:2006.03492v1 fatcat:xkypayfysfh53kuucodcxzjety

RTL3D: real-time LIDAR-based 3D object detection with sparse CNN

Lin Yan, Kai Liu, Evgeny Belyaev, Meiyu Duan
2020 IET Computer Vision  
To find a balance between speed and accuracy for 3D object detection from LIDAR, authors propose RTL3D, a computationally efficient Real-time LIDAR-based 3D detector.  ...  LIDAR (light detection and ranging) based real-time 3D perception is crucial for applications such as autonomous driving.  ...  In addition, some multi-modal fusion methods are reviewed. 3D object detection from 2D image Monocular-based 3D detection has a similar framework as 2D.  ... 
doi:10.1049/iet-cvi.2019.0508 fatcat:wvmstqmmrbdhleqwpjue3u6vza

A system for 3D texture-based probabilistic object recognition and its applications

Marcin Grzegorzek
2009 Pattern Analysis and Applications  
Rather, the objects are described by local feature vectors computed directly from image pixel values using the wavelet transform. Both gray level and colour images can be processed.  ...  This article presents a system for texture-based probabilistic classification and localisation of threedimensional objects in two-dimensional digital images and discusses selected applications.  ...  [1] proposes an algorithm for multi-class shape detection in the sense of recognising and localising instances from multiple shape classes.  ... 
doi:10.1007/s10044-009-0163-0 fatcat:2sckprhi3zaqtpri7ang7jzpwm

3D Room Geometry Reconstruction Using Audio-Visual Sensors

Hansung Kim, Luca Remaggi, Philip JB Jackson, Filippo Maria Fazi, Adrian Hilton
2017 2017 International Conference on 3D Vision (3DV)  
Existing vision-based 3D reconstruction methods are not applicable for scenes with transparent or reflective objects such as windows and mirrors.  ...  Depth information of the scene is recovered by stereo matching from the captured images and estimation of major acoustic reflector locations from the sound.  ...  Details about the data underlying this work are available from: http://  ... 
doi:10.1109/3dv.2017.00076 dblp:conf/3dim/KimRJFH17 fatcat:e3ssonjk2rcsdcu7mpkymhkckm

A Review on Deep Learning Techniques for 3D Sensed Data Classification

David Griffiths, Jan Boehm
2019 Remote Sensing  
We review the current main approaches, including RGB-D, multi-view, volumetric and fully end-to-end architecture designs. Datasets for each category are documented and explained.  ...  Over the past decade deep learning has driven progress in 2D image understanding.  ...  Song and Xiao [72] proposed a method to take depth images from low cost sensors and perform amodal 3D object classification and localisation (object detection).  ... 
doi:10.3390/rs11121499 fatcat:vg7bme6guzddleocgsm7soqaxq

Local Wavelet Features for Statistical Object Classification and Localisation

Marcin Grzegorzek, Sorin Sav, Ebroul Izquierdo, Noel O'Connor
2009 IEEE Multimedia  
This article presents a system for texture-based probabilistic classification and localisation of 3D objects in 2D digital images and discusses selected applications.  ...  Experiments carried out on a real dataset of over 40000 images demonstrate the robustness of the system in terms of classification and localisation accuracy.  ...  Each image contains between one and three objects from the 3D-REAL-ENV database pictured in Figure 3 .  ... 
doi:10.1109/mmul.2009.67 fatcat:uz6uo6dk6vho7jshe5lh43g464

Deep Learning for Real-Time 3D Multi-Object Detection, Localisation, and Tracking: Application to Smart Mobility

Antoine Mauri, Redouane Khemmar, Benoit Decoux, Nicolas Ragot, Romain Rossi, Rim Trabelsi, Rémi Boutteau, Jean-Yves Ertaud, Xavier Savatier
2020 Sensors  
In core computer vision tasks, we have witnessed significant advances in object detection, localisation and tracking.  ...  In this paper, our objective is to develop a deep learning multi object detection and tracking technique applied to road smart mobility.  ...  In this context, the work presented in this paper deals with the task of multi-object detection, localisation and tracking for smart road mobility.  ... 
doi:10.3390/s20020532 pmid:31963641 pmcid:PMC7014509 fatcat:rzrejejeqzcwdeldiazcftwzii
« Previous Showing results 1 — 15 out of 11,903 results