Accurate Multiple View 3D Reconstruction Using Patch-Based Stereo for Large-Scale Scenes

Shen, Shuhan

doi:10.1109/tip.2013.2237921

Cited by 247 publications

(107 citation statements)

References 32 publications

Supporting

Mentioning

107

Contrasting

Order By: Relevance

“…Accurate and precise geological measurements thus require proper scaling. The 3D models have been reconstructed using an adapted 3D reconstruction procedure consisting of multiple open-source solutions (OpenMVG (Moulon et al, 2013), OpenMVS (Shen, 2013;Jancosek and Pajdla, 2014), MVS-Texturing (Waechter et al, 2014)) as described in (Hernández et al, 2016). Figure 16 depicts one such model, named FPA, which has been reconstructed from a total of 218 images with the resolution of 1920 × 1080.…”

Section: Real Scenariomentioning

confidence: 99%

Automatic scale estimation of structure from motion based 3D models using laser scalers in underwater scenarios

Istenič

Gracias

Arnaubec

et al. 2020

ISPRS Journal of Photogrammetry and Remote Sensing

View full text Add to dashboard Cite

A B S T R A C TRecent advances in structure-from-motion techniques are enabling many scientific fields to benefit from the routine creation of detailed 3D models. However, for a large number of applications, only a single camera is available for the image acquisition, due to cost or space constraints in the survey platforms. Monocular structure-from-motion raises the issue of properly estimating the scale of the 3D models, in order to later use those models for metrology. The scale can be determined from the presence of visible objects of known dimensions, or from information on the magnitude of the camera motion provided by other sensors, such as GPS.This paper addresses the problem of accurately scaling 3D models created from monocular cameras in GPS-denied environments, such as in underwater applications. Motivated by the common availability of underwater laser scalers, we present two novel approaches which are suitable for different laser scaler configurations. A fully-calibrated method enables the use of arbitrary laser setups, while a partially-calibrated method reduces the need for calibration by only assuming parallelism on the laser beams, with no constraints on the camera. The proposed methods have several advantages with respect to the existing methods. By using the known geometry of the scene expressed by the 3D model, along with some parameters of the laser scaler geometry, the need for laser alignment with the optical axis of the camera is removed. Furthermore, the extremely error-prone manual identification of image points on the 3D model, currently required in image-scaling methods, is eliminated as well.The performance of the methods and their applicability was evaluated on both data generated from a realistic 3D model and data collected during an oceanographic cruise in 2017. Three separate laser configurations have been tested, encompassing nearly all possible laser setups, to evaluate the effects of terrain roughness, noise, camera perspective angle and camera-scene distance on the final estimates of scale. In the real scenario, the computation of 6 independent model scale estimates using our fullycalibrated approach, produced values with standard deviation of 0.3%. By comparing the values to the only possible method usable for this dataset, we showed that the consistency of scales obtained for individual lasers is much higher for our approach (0.6% compared to 4%).

show abstract

Section: Real Scenariomentioning

confidence: 99%

Automatic scale estimation of structure from motion based 3D models using laser scalers in underwater scenarios

Istenič

Gracias

Arnaubec

et al. 2020

ISPRS Journal of Photogrammetry and Remote Sensing

View full text Add to dashboard Cite

show abstract

“…To verify the effectiveness of our proposed cost computation method, we used six pairs of stereo images with ground truth: Aloe, Baby1, Bowling1, Cloth1, Flowerpots, and Rocks 1, provided by Middlebury 2006 datasets [32] which have three illuminations (1,2,3) and three different exposure settings (0, 1, 2). Figure 7 shows the left image of the aloe data under three different illuminations (no exposure variation) and with three different exposure settings (no illumination change).…”

Section: Evaluation Of the Robustness To The Illumination And Exposurmentioning

confidence: 99%

“…Dense stereo matching is a significant research topic in the field of photogrammetry and computer vision, greatly benefiting applications like 3D reconstruction, DSM (Digital Surface Model) production, visual reality and autonomous vehicles [1][2][3][4]. A large number of efficient stereo matching algorithms have been developed in recent years, but it is still a challenging task to handle the stereo matching problem in occluded, textureless and discontinuous regions.…”

Section: Introductionmentioning

confidence: 99%

Improved Cost Computation and Adaptive Shape Guided Filter for Local Stereo Matching of Low Texture Stereo Images

et al. 2020

View full text Add to dashboard Cite

Dense stereo matching has been widely used in photogrammetry and computer vision applications. Even though it has a long research history, dense stereo matching is still challenging for occluded, textureless and discontinuous regions. This paper proposed an efficient and effective matching cost measurement and an adaptive shape guided filter-based matching cost aggregation method to improve the stereo matching performance for large textureless regions. At first, an efficient matching cost function combining enhanced image gradient-based matching cost and improved census transform-based matching cost is introduced. This proposed matching cost function is robust against radiometric variations and textureless regions. Following this, an adaptive shape cross-based window is constructed for each pixel and a modified guided filter based on this adaptive shape window is implemented for cost aggregation. The final disparity map is obtained after disparity selection and multiple steps disparity refinement. Experiments were conducted on the Middlebury benchmark dataset to evaluate the effectiveness of the proposed cost measurement and cost aggregation strategy. The experimental results demonstrated that the average matching error rate on Middlebury standard image pairs is 9.40%. Compared with the traditional guided filter-based stereo matching method, the proposed method achieved a better matching result in textureless regions.

show abstract

“…The software used a structure from motion library (OpenSfM, Mapillary AB, Malmö, Sweden) to perform feature extraction matching and 3-D point cloud computation. Construction of dense point clouds was based on the multi-view stereo algorithm of Shen [29]. Geographic coordinates from 81 ground control points (GCPs) evenly distributed across the field area were surveyed with an RTK-GPS receiver and were incorporated into the photogrammetry calculations for georeferencing purposes.…”

Section: Uas Imagesmentioning

confidence: 99%

Comparing Nadir and Multi-Angle View Sensor Technologies for Measuring in-Field Plant Height of Upland Cotton

et al. 2019

View full text Add to dashboard Cite

Plant height is a morphological characteristic of plant growth that is a useful indicator of plant stress resulting from water and nutrient deficit. While height is a relatively simple trait, it can be difficult to measure accurately, especially in crops with complex canopy architectures like cotton. This paper describes the deployment of four nadir view ultrasonic transducers (UTs), two light detection and ranging (LiDAR) systems, and an unmanned aerial system (UAS) with a digital color camera to characterize plant height in an upland cotton breeding trial. The comparison of the UTs with manual measurements demonstrated that the Honeywell and Pepperl+Fuchs sensors provided more precise estimates of plant height than the MaxSonar and db3 Pulsar sensors. Performance of the multi-angle view LiDAR and UAS technologies demonstrated that the UAS derived 3-D point clouds had stronger correlations (0.980) with the UTs than the proximal LiDAR sensors. As manual measurements require increased time and labor in large breeding trials and are prone to human error reducing repeatability, UT and UAS technologies are an efficient and effective means of characterizing cotton plant height.height is often positively correlated with fiber yield and water-use efficiency when cotton is grown with low soil moisture conditions [9,10]. These studies indicate that plant height is a useful phenotype for cotton breeders to identify high yielding lines, particularly under less-than-optimal environmental conditions. Unfortunately, manual measurements of plant height are labor intensive, prone to human error, and provide data for a limited number of plants. These factors lead to extreme inefficiencies for large scale breeding programs, in which thousands of such measurements must be collected.Field-based high-throughput phenotyping is a novel approach for increasing plant breeding efficiency by using proximal sensors and imagers on terrestrial or aerial platforms to generate useful phenotypic data for plant characterization. The power of high-throughput phenotyping (HTP) is the ability to characterize large populations in both time and space, which improves trait capture of dynamic plant responses to environmental conditions. Over the last 10 years, a multitude of HTP platforms and sensor packages have been developed and evaluated in various agronomic crops [11][12][13]. Platforms and sensors used for plant height evaluation include ultrasonic transducers mounted on high-clearance tractors or field carts [10,14-16], light detection and ranging (LiDAR) systems mounted on tractors and rails [13,[17][18][19][20], and compact digital color cameras mounted on either unmanned aerial systems (UAS) or terrestrial platforms [16,19,21,22].The fundamentals of ultrasonic transducer (UT) sensors are based on the time-of-flight (TOF) principle. These devices use a transmitter to send ultrasonic pulses toward an object at frequencies higher than 20 kHz while a receiver registers the pulses reflected back from the object. Distance is calculated as half of th...

show abstract

Accurate Multiple View 3D Reconstruction Using Patch-Based Stereo for Large-Scale Scenes

Cited by 247 publications

References 32 publications

Automatic scale estimation of structure from motion based 3D models using laser scalers in underwater scenarios

Automatic scale estimation of structure from motion based 3D models using laser scalers in underwater scenarios

Improved Cost Computation and Adaptive Shape Guided Filter for Local Stereo Matching of Low Texture Stereo Images

Comparing Nadir and Multi-Angle View Sensor Technologies for Measuring in-Field Plant Height of Upland Cotton

Contact Info

Product

Resources

About