High‐Quality Stereo Depth Map Generation Using Infrared Pattern Projection

Jeong, Jae-chan; Chang, Jung Min; Lim, Eul-Gyun; Choi, Seung Min; Yoon, Kuk-Jin; Cho, Jae-il

doi:10.4218/etrij.13.2013.0052

Cited by 14 publications

(16 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It then sorts all vectors according to a measured angle and filters out any misrecognized vectors in the process. Finally, using the outer-most points of an object's respective feature points, it then employs a further algorithm (convex hull) to extract the outlines of all existing objects [27]- [29].…”

Section: Local Mapping Using Corrected Points Of An Objectmentioning

confidence: 99%

Multi-robot Mapping Using Omnidirectional-Vision SLAM Based on Fisheye Images

Choi

Kwon

Lee

et al. 2014

ETRI J

View full text Add to dashboard Cite

This paper proposes a global mapping algorithm for multiple robots from an omnidirectional-vision simultaneous localization and mapping (SLAM) approach based on an object extraction method using LucasKanade optical flow motion detection and images obtained through fisheye lenses mounted on robots. The multi-robot mapping algorithm draws a global map by using map data obtained from all of the individual robots. Global mapping takes a long time to process because it exchanges map data from individual robots while searching all areas. An omnidirectional image sensor has many advantages for object detection and mapping because it can measure all information around a robot simultaneously. The process calculations of the correction algorithm are improved over existing methods by correcting only the object's feature points. The proposed algorithm has two steps: first, a local map is created based on an omnidirectional-vision SLAM approach for individual robots. Second, a global map is generated by merging individual maps from multiple robots. The reliability of the proposed mapping algorithm is verified through a comparison of maps based on the proposed algorithm and real maps.

show abstract

Section: Local Mapping Using Corrected Points Of An Objectmentioning

confidence: 99%

Multi-robot Mapping Using Omnidirectional-Vision SLAM Based on Fisheye Images

Choi

Kwon

Lee

et al. 2014

ETRI J

View full text Add to dashboard Cite

show abstract

“…In particular, to understand outdoor scenes, the necessity of 3D sensors has been increasing since the success of the DARPA Grand and Urban Challenges as well as the appearance of the Google self driving cars [1]. The 3D sensors provide spatial information, such as 3D point clouds, and camera based sensors also give depth and color information in 2D images [2]. The 3D information helps robots or intelligent vehicles to understand the real world more precisely.…”

Section: Introductionmentioning

confidence: 99%

Quadtree sampling-based superpixels for 3D range data

Park

Choi

2014

2014 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

3D range sensors are currently being used in various fields. Creating 3D range sensors requires various techniques, such as object detection, tracking, classification, 3D SLAM, etc. For the pre-processing step, superpixels can improve the performance of these techniques. This paper proposes a novel over-segmentation algorithm, known as superpixels, for 3D outdoor urban range data. Superpixels are generated with three steps: boundary extraction using a surface change score and sensor models, initial cluster seeding using a quadtree decomposition, and iterative clustering, which adapts a kmeans clustering approach with limited search size in the quadtree dimension. The proposed algorithm produces adaptive superpixel sizes that take into account surface and object border information. This reduces memory size more than regular grid methods and represents small objects well with adaptable pixel sizes. The algorithm is verified using the publicly available Velodyne dataset and the manually annotated ground truth. A comparison with the conventional algorithm is also presented.

show abstract

“…Many studies on stereo vision for obtaining three-dimensional information have been conducted [1]- [4]. Since the release of Microsoft Kinect (2009), in particular, researchers are paying more attention to applications that use 3D depth sensors.…”

Section: Introductionmentioning

confidence: 99%

“…However, applications using 3D depth information have trouble detecting long and thin objects, such as a human finger, at a distance of more than 3 m owing to the performance margins of 3D depth sensors. To overcome this limitation, Jeong and others proposed a stereo matching system using a time-division pattern projection [1], and Chang and others modified this system for implementation in four field-programmable gate arrays (FPGAs) for real-time processing [2]. Chang and others stated that their systems can calculate the depth for a 1,280 × 720 resolution image at 60 fps in an indoor environment [2].…”

Section: Introductionmentioning

confidence: 99%

Implementation of Real-Time Post-Processing for High-Quality Stereo Vision

Choi¹,

Jeong²,

Chang³

et al. 2015

ETRI J

Self Cite

View full text Add to dashboard Cite

We propose a novel post‐processing algorithm and its very‐large‐scale integration architecture that simultaneously uses the passive and active stereo vision information to improve the reliability of the three‐dimensional disparity in a hybrid stereo vision system. The proposed architecture consists of four steps — left‐right consistency checking, semi‐2D hole filling, a tiny adaptive variance checking, and a 2D weighted median filter. The experimental results show that the error rate of the proposed algorithm (5.77%) is less than that of a raw disparity (10.12%) for a real‐world camera image having a 1,280×720 resolution and maximum disparity of 256. Moreover, for the famous Middlebury stereo image sets, the proposed algorithm's error rate (8.30%) is also less than that of the raw disparity (13.7%). The proposed architecture is implemented on a single commercial field‐programmable gate array using only 13.01% of slice resources, which achieves a rate of 60 fps for 1,280×720 stereo images with a disparity range of 256.

show abstract

High‐Quality Stereo Depth Map Generation Using Infrared Pattern Projection

Cited by 14 publications

References 18 publications

Multi-robot Mapping Using Omnidirectional-Vision SLAM Based on Fisheye Images

Multi-robot Mapping Using Omnidirectional-Vision SLAM Based on Fisheye Images

Quadtree sampling-based superpixels for 3D range data

Implementation of Real-Time Post-Processing for High-Quality Stereo Vision

Contact Info

Product

Resources

About