Wrong Depth Research Articles

Geometric Prior-Guided Self-Supervised Learning for Multi-View Stereo

Recently, self-supervised multi-view stereo (MVS) methods, which are dependent primarily on optimizing networks using photometric consistency, have made clear progress. However, the difference in lighting between different views and reflective objects in the scene can make photometric consistency unreliable. To address this issue, a geometric prior-guided multi-view stereo (GP-MVS) for self-supervised learning is proposed, which exploits the geometric prior from the input data to obtain high-quality depth pseudo-labels. Specifically, two types of pseudo-labels for self-supervised MVS are proposed, based on the structure-from-motion (SfM) and traditional MVS methods. One converts the sparse points of SfM into sparse depth maps and combines the depth maps with spatial smoothness constraints to obtain a sparse prior loss. The other generates initial depth maps for semi-dense depth pseudo-labels using the traditional MVS, and applies a geometric consistency check to filter the wrong depth in the initial depth maps. We conducted extensive experiments on the DTU and Tanks and Temples datasets, which demonstrate that our method achieves state-of-the-art performance compared to existing unsupervised/self-supervised approaches, and even performs on par with traditional and supervised approaches.

A dense map optimization method based on common-view geometry

This paper focuses on noisy points filtering in dynamic reconstruction. RGBD cameras can acquire depth information directly, which is an effective way to obtain a global point cloud. However, the inevitable error leads to erroneous rebuilding, which is mainly caused by wrong depth information and dynamic disturbance. Different from algebraic methods such as median filter, a geometric method is proposed based on common-view geometry. Noisy points generate more intersections with other projection lights, which are used to distinguish incorrect points. Considering the huge amount of computation, a sampling method is adopted to avoid unnecessary calculations and a cuboid method to simplify the calculation. We also employ a hybrid strategy with two different standards to increase the robustness when filtering. By calculating the geometry relationship effectively and robustly, we get good results in removing map points with incorrectly measured depth and caused by dynamic disturbance. Compared with the algebraic methods, the proposed method can also remove dynamic objects. The performance is verified in three popular public datasets.

Control of technological processes and reduction of gas dynamic hazard on the basis of using the acoustic afteraction

The objective of the work was to develop new methods for controlling technological processes during driving development headings and mining coal in stope faces. Accidents happened while performing of underground mining operations are often caused by the so-called “human factor”: miners do not always accurately follow the safety regulations, namely, the roadways are not supported in due time, the gas-drainage boreholes are drilled to the wrong depth, duration of the fluid injection at hydraulic loosening is cut, etc. In real conditions, it is difficult to control all the technological processes occurred underground. Therefore, the purpose of the study was to develop recommendations for controlling technological processes in the mines in order to reduce the risks of gas-dynamic phenomenon occurrence. In this work, the authors present further development of the most rational known approach to the control of technological processes consisted in assessing of gas-dynamic hazard by the method of seismoacoustic aftereffect. The nature of changes in acoustic activity depending on the type of winning operations was analyzed. The clearer differentiated assessment of the stress-strain state of the coal seam face area was obtained. The study of the acoustic aftereffect during the mining of a coal seam by different technological methods made it possible to assess the technology of winning operations from the point of view of ensuring safe working conditions. The block diagram was built, on the basis of which an algorithm of performing operational control of technological processes in development headings and stope faces was developed, including operations in prone-to-outburst seams. The basic requirements for the correction of existing software for the classification and recognition of technological processes by their acoustic aftereffect (trace) in the records of the predicting systems carried out during mining operations in the mine were formulated. These recommendations for the control of technological processes were proposed the Pokrovskoe Mine for implementation. The proposed recommendations improve accuracy of the forecasting of gas-dynamic hazards during mining operations, make it possible to control technological processes as well as simplify the investigation procedure for establish the causes of a critical situation in the case of an accident. The implementation of the recommendations will also make it possible to assess a gas-dynamic hazard during stoppages for repair and other auxiliary work, to predict geological disturbances ahead of the moving face, to determine the size of the unloading zone of the bottom area of the coal seam, and to assess the effectiveness of measures to prevent gas-dynamic phenomena. Keywords: acoustic aftereffect, technological processes, gas-dynamic phenomenon, safety of mining operations.

Iterative Refinement of Uniformly Focused Image Set for Accurate Depth from Focus

Estimating the 3D shape of a scene from differently focused set of images has been a practical approach for 3D reconstruction with color cameras. However, reconstructed depth with existing depth from focus (DFF) methods still suffer from poor quality with textureless and object boundary regions. In this paper, we propose an improved depth estimation based on depth from focus iteratively refining 3D shape from uniformly focused image set (UFIS). We investigated the appearance changes in spatial and frequency domains in iterative manner. In order to achieve sub-frame accuracy in depth estimation, optimal location of focused frame in DFF is estimated by fitting a polynomial curve on the dissimilarity measurements. In order to avoid wrong depth values on texture-less regions we propose to build a confidence map and use it to identify erroneous depth estimations. We evaluated our method on public and our own datasets obtained from different types of devices, such as smartphones, medical, and normal color cameras. Quantitative and qualitative evaluations on various test image sets show promising performance of the proposed method in depth estimation.

Guided optimization framework for the fusion of time-of-flight with stereo depth

The fusion of depth acquired actively with the depth estimated passively proved its significance as an improvement strategy for gaining depth. This combination allows us to benefit from two sources of modalities such that they complement each other. To fuse two sensor data into a more accurate depth map, we must consider the limitations of active sensing such as low lateral resolution while combining it with a passive depth map. We present an approach for the fusion of active time-of-flight depth and passive stereo depth in an accurate way. We propose a multimodal sensor fusion strategy that is based on a weighted energy optimization problem. The weights are generated as a result of combining the edge information from a texture map and active and passive depth maps. The objective evaluation of our fusion algorithm shows an improved accuracy of the generated depth map in comparison with the depth map of every single modality and with the results of other fusion methods. Additionally, a visual comparison of our result shows a better recovery on the edges considering the wrong depth values estimated in passive stereo. Moreover, the left and right consistency check on the result illustrates the ability of our approach to consistently fuse sensors.

Reparation of transparency for high-quality three-dimensional reconstruction using a single RGB-D sensor

Consumer-grade range cameras are widely used in three-dimensional reconstruction. However, the resolution and stability limit the quality of reconstruction, especially with transparent objects. A method is proposed to reconstruct the transparency while improving the reconstruction quality of the indoor scene with a single RGB-D sensor. We propose the method to localize the transparent regions from zero depth and wrong depth. The lost surface of transparency is recovered by modeling the statistics of zero depth, variance, and residual error of signed distance function (SDF) with depth data fusion. The camera pose is first initialized by the error minimization of depth map on the SDF and k-color-frame constraint. The pose then is optimized by the penalized coefficient function, which lowers the weight of voxels with higher SDF error. The method is proved to be valid in localizing the transparent objects and can achieve a more robust camera pose under a complex background.

Temperature Transients Affect Reservoir–Pressure Estimation During Well Tests: Case Study and Model

Summary This paper discusses how the pressure measured by the downhole gauges during a drillstem test (DST) is affected not only by hydrostatic, friction, or kinetic-energy pressure losses, but also by changes in temperature. In oil wells, the bottomhole temperature is decreasing when the pressure is building up after a flow period. The change in temperature is inducing a change in the wellbore-fluid density, itself affecting the bottomhole pressure. Most numerical-simulators modeling flow in pipes do not take into account this effect, because it is not significant for low- to moderate-productivity wells. In high-productivity wells, however, the temperature change can affect dramatically the bottomhole pressure measured above the reservoir, and the pressure can even decrease at the end of a static period. A mathematical model is presented to compute the pressure drop caused by the nonisothermal effects, and it can be used to correct the measurements. Field examples are discussed to illustrate this effect, and recommendations are made on how new technology can improve pressure measurements during well tests. To measure the effects on the pressure measurements caused by the position of the gauges in the DST string, two gauge carriers were run in some well tests in deepwater wells offshore Brazil, one at the level of the tester valve and one below the packer, as deep as possible in the test string. The gauges could transmit data with a wireless telemetry system that allowed the operator to monitor in real time the pressure changes and the difference between the two sets of gauges. In high-permeability wells, using the uncorrected pressure measured with a gauge 50 to 100 m above the reservoir, as is typically the case, can result in a wrong productivity index (PI) and, at times, in uninterpretable results. Locating pressure gauges as close to the reservoir as possible is crucial in high-permeability reservoirs, to minimize the measurement errors, but also to avoid misinterpretations when an obstruction forms below the tester valve. With the new wireless telemetry systems, it is possible to monitor in real time gauges mounted below the packer, much closer to the perforations than the standard gauges placed at the level of the tester valve, thus avoiding making wrong decisions during the test. A long pressure-transient test could be uninterpretable because the gauges are at the wrong depth. Ideally, the pressure gauges should be placed in front of the perforations, but it is not always possible. Therefore, nonisothermal effects on pressure measurements during a well test need to be clearly understood and taken into account when designing and interpreting well tests, particularly in high-productivity reservoirs.

The stochastic highlighting of polygon edges in the transparent visualization of large-scale polygon meshes: application to visualizing a high-energy elementary particle detector

ABSTRACTIn this paper, we propose a new stochastic algorithm that supports high-quality transparent visualization of line objects and apply it to polygon edges of complex three-dimensional (3D) polygon models. We also apply it to realizing transparent wireframe visualization that greatly improves comprehensibility of the conventional method. In our algorithm, the line opacity is recognized as the probability that a pixel on the image plane becomes the projected line colour. Our stochastic algorithm successfully makes a depth sort of rendering primitives, which often causes rendering artefacts and a wrong depth feel in conventional non-stochastic transparent rendering, unnecessary. We demonstrate our method using 3D models of the ATLAS detector, which is the world's largest high-energy elementary particle detector at CERN laboratory (EU) and a complex car engine. We also demonstrate the transparent wireframe visualization using the simulation data, which consist of many line objects, acquired for the ATLAS detector.

Fusing Depth and Silhouette for Scanning Transparent Object with RGB-D Sensor

3D reconstruction based on structured light or laser scan has been widely used in industrial measurement, robot navigation, and virtual reality. However, most modern range sensors fail to scan transparent objects and some other special materials, of which the surface cannot reflect back the accurate depth because of the absorption and refraction of light. In this paper, we fuse the depth and silhouette information from an RGB-D sensor (Kinect v1) to recover the lost surface of transparent objects. Our system is divided into two parts. First, we utilize the zero and wrong depth led by transparent materials from multiple views to search for the 3D region which contains the transparent object. Then, based on shape from silhouette technology, we recover the 3D model by visual hull within these noisy regions. Joint Grabcut segmentation is operated on multiple color images to extract the silhouette. The initial constraint for Grabcut is automatically determined. Experiments validate that our approach can improve the 3D model of transparent object in real-world scene. Our system is time-saving, robust, and without any interactive operation throughout the process.

Depth Estimates of Buried Utility Systems Using the GPR Method: Studies at the IAG/USP Geophysics Test Site

Identifying underground utilities and predicting their depth are fundamental when it comes to civil engineering excavations, for example, to install or repair water, sewer, gas, electric systems and others. The accidental rupture of these systems can lead to unplanned repair costs, delays in completing the service, and risk injury or death of workers. One way to detect underground utilities is using the GPR-Ground Penetrating Radar geophysical method. To estimate depth, the travel time (two-way travel time) information provided by a radargram is used in conjunction with ground wave velocity, which depends on the dielectric constant of materials, where it is usually assumed to be constant for the area under investigation. This procedure provides satisfactory results in most cases. However, wrong depth estimates can result in damage to public utilities, rupturing pipes, cutting lines and so on. These cases occur mainly in areas that have a marked variation of water content and/or soil lithology, thus greater care is required to determine the depth of the targets. The present work demonstrates how the interval velocity of Dix (1955) can be applied in radargram to estimate the depth of underground utilities compared to the conventional technique of constant velocity applied to the same data set. To accomplish this, synthetic and real GPR data were used to verify the applicability of the interval velocity technique and to determine the accuracy of the depth estimates obtained. The studies were carried out at the IAG/USP test site, a controlled environment, where metallic drums are buried in known positions and depths allowing the comparison of real to estimated depths. Numerical studies were also carried out aiming to simulate the real environment with variation of dielectric constant in depth and to validate the results with real data. The results showed that the depths of the targets were estimated more accurately by means of the interval velocity technique in contrast to the constant velocity technique, minimizing the risks of accidents during excavation.

Adaptive High Dynamic Range for Time-of-Flight Cameras

The limited dynamic range of conventional digital cameras is a well-known problem. A common solution is to apply high-dynamic-range (HDR) techniques over several images acquired using different exposure times. Time-of-flight (ToF) cameras using a photonic mixer device (PMD) are not an exception, since the dynamic range of its dual pixels is also limited. Furthermore, in this case, the saturation of the pixel channels leads to wrong depth measurements. An appropriate solution is the suppression of background illumination (SBI) system designed by the PMD. This system actually extends the dynamic range of the camera by hardware, but it also introduces noise when activated. In this paper, we present an adaptive HDR (AHDR) solution to the problem for the ToF case that overcomes the limited dynamic range of the system, allowing sensing along a theoretically infinite dynamic range with the only limitations of the power of the illumination system and the decay of the SNR with higher distances or lower illumination intensities. Our method is able to detect and segment relevant scene regions responsible for unexpected saturation, i.e., close foreground objects, from the rest of the scene and adjust the exposure times of the acquired images considering them. The results show a reduction in detail losses and a higher SNR in the AHDR raw images, with respect to single acquisitions. This results in a dramatic depth error reduction and effective axial resolution improvement in critical areas, while keeping a high frame rate. In addition, the SBI-related noise is eliminated.

Direction hole-filling method for a 3D view generator

Depth image-based rendering DIBR technology is an approach to creating a virtual 3D image from one single 2D image. A desired view can be synthesised at the receiver side using depth images to make transmission and storage efficient. While this technique has many advantages, one of the key challenges is how to fill the holes caused by disocclusion regions and wrong depth values in the warped left/right images. A common means to alleviate the sizes and the number of holes is to smooth the depth image. But smoothing results in geometric distortions and degrades the depth image quality. This study proposes a hole-filling method based on the oriented texture direction. Parallax correction is first implemented to mitigate the wrong depth values. Texture directional information is then probed in the background pixels where holes take place after warping. Next, in the warped image, holes are filled according to their directions. Experimental results showed that this algorithm preserves the depth information and greatly reduces the amount of geometric distortion.

Segmentation-based view synthesis for multi-view video plus depth

In this paper, we present a novel view synthesis algorithm for three-dimensional video. The proposed algorithm is based on segmentation using multi-level thresholding method. Recently, numerous techniques have been suggested which use a 2-D color image and the per-pixel depth map of the scene to create virtual views of the scene from any viewing position. However, inaccuracy in the depth maps cause annoying visual artifacts in depth-based view synthesis. In the proposed method, the depth maps are first preprocessed to avoid the errors caused by wrong depth values. Then, the color images are segmented according to the depth values and the regions belonging to different segments are warped independently. To further enhance the quality of the synthesized views, a multi-level thresholding based ghost removal algorithm and a novel hole filling algorithm have been proposed. Experimental results show that the proposed methods achieve an average PSNR gain of 0.98 dB for the multi-view test sequences and also improve the subjective quality of the synthesized views.

Subsurface offset behaviour in velocity analysis with extended reflectivity images

ABSTRACTMigration velocity analysis with the constant‐density acoustic wave equation can be accomplished by the focusing of extended migration images, obtained by introducing a subsurface shift in the imaging condition. A reflector in a wrong velocity model will show up as a curve in the extended image. In the correct model, it should collapse to a point. The usual approach to obtain a focused image involves a cost functional that penalizes energy in the extended image at non‐zero shift. Its minimization by a gradient‐based method should then produce the correct velocity model. Here, asymptotic analysis and numerical examples show that this method may be too sensitive to amplitude peaks at large shifts at the wrong depth and to artefacts. A more robust alternative is proposed that can be interpreted as a generalization of stack power and maximizes the energy at zero‐subsurface shift. A real‐data example is included.

Identification and Reconstruction of Single-Pixel Incomplete Charge Collection Events

Incomplete charge collection in pixelated CdZnTe detectors with a common anode steering grid has been studied using an advanced digital ASIC readout system, which is capable of sampling and digitizing the preamplifier output signals of an array of 11×11 pixelated anodes as well as the cathode and grid electrodes. Due to insufficient steering, only a fraction of the electron cloud may be collected by anode pixels for some events. These events are called incomplete charge collection events, and they tend to take place on edge pixels. Since the current single-pixel event reconstruction algorithm is based on the anode signal amplitude and the cathode-to-anode signal amplitude ratio (C/A), these events will be reconstructed to the wrong depth and energy bins. An algorithm has been developed to reconstruct these special events based on the system response function. Experiments were performed on a pixelated 20×20×15-mm3 CdZnTe detector with a common grid between pixels manufactured by Redlen Technologies. With this algorithm, both the interaction depth and energy deposition can be reconstructed correctly in the absence of a reliable anode amplitude. This reconstruction algorithm improves the single-pixel photopeak counts by close to 10%. The electron cloud size for 662 keV was estimated with a square-shaped model based on the improvement of photopeak efficiency and showed consistent results with Geant4 simulations.

A stereoscopic video conversion scheme based on spatio-temporal analysis of MPEG videos

Abstract In this article, an automatic stereoscopic video conversion scheme which accepts MPEG-encoded videos as input is proposed. Our scheme is depth-based, relying on spatio-temporal analysis of the decoded video data to yield depth perception cues, such as temporal motion and spatial contrast, which reflect the relative depths between the foreground and the background areas. Our scheme is shot-adaptive, demanding that shot change detection and shot classification be performed for tuning of algorithm or parameters that are used for depth cue combination. The above-mentioned depth estimation is initially block-based, followed by a locally adaptive joint trilateral upsampling algorithm to reduce the computing load significantly. A recursive temporal filter is used to reduce the possible depth fluctuations (and also artifacts in the synthesized images) resulting from wrong depth estimations. The traditional Depth-Image-Based-Rendering algorithm is used to synthesize the left- and right-view frames for 3D display. Subjective tests show that videos converted by our scheme provide comparable perceived depth and visual quality with those converted from the depth data calculated by stereo vision techniques. Also, our scheme is shown to outperform the well-known TriDef software in terms of human’s perceived 3D depth. Based on the implementation by using “OpenMP” parallel programming model, our scheme is capable of executing in real-time on a multi-core CPU platform.

Hierarchical Hole-Filling For Depth-Based View Synthesis in FTV and 3D Video

Three-dimensional television (3DTV) is believed to be the future of television broadcasting that would replace current 2D HDTV technology. Future 3DTV would bring a more life-like and visually immersive home entertainment experience, in which users will have the freedom to navigate through the scene to choose a different viewpoint. A desired view can be synthesized at the receiver side using depth image-based rendering (DIBR). While this approach has many advantages, one of the key challenges in DIBR is how to fill the holes caused by disocclusion regions and wrong depth values. In this paper, we propose two new approaches for disocclusion removal in DIBR. Both approaches namely hierarchical hole-filling (HHF) and depth adaptive hierarchical hole-filling eliminate the need for any smoothing or filtering of the depth map. Both techniques use a pyramid-like approach to estimate the hole pixels from lower resolution estimates of the 3D wrapped image. The lower resolution estimates involve a pseudo zero canceling plus Gaussian filtering of the wrapped image. The depth adaptive HHF incorporates the depth information to produce a higher resolution rendering around previously occluded areas. Experimental results show that HHF and depth adaptive HHF yield virtual images and stereoscopic videos that are free of geometric distortions and a better rendering quality both subjectively and objectively than traditional hole-filling approaches.

Effects of the amount of monocular shape information on stereo scaling problem

AbstractThe perception of a three‐dimensional shape from binocular disparity depends on the estimation of viewing distance. Therefore, in the absence of good distance information, an observer misestimates the viewing distance and perceives a wrong depth for an object. While motion depth cues can theoretically overcome this problem, past studies have reported inconsistent results. We examined whether differences in the amount of monocular shape information between the stimuli used in these past studies can resolve the discrepancy. The amount of information was manipulated by changing the presence of object boundaries, dot density, and rotation angles. The results indicate the effect of the amount of monocular shape information and thus suggest its significance in accounting for the discrepancies among past studies. It should also be noted that shape factors interact with viewing distance. Motion cues cannot correct distortion at near viewing distance. This suggests that the cue that gives the larger depth dominates in determining perceived shape.

On methods to select solutions in Euler deconvolution of gravitation and gradient measurements

Euler deconvolution is a standard technique in geophysical exploration to determine the depth of a subsurface mass density or magnetic anomaly from surface measurements of the corresponding field. Because the solution is unstable, some selection criterion must be employed to eliminate probably wrong depth estimates. An ad hoc decision criterion often used by geophysicists is compared here to a probabilistically founded test of the estimates using the Fisher F-distribution. It is shown that this test performs quite badly compared to the ad hoc criterion. However, a probabilistic foundation for the ad hoc criterion is also developed here using the coefficient of variation. Simulations are used to verify that it offers a methodological approach to selecting estimates based on a probabilistic relationship to their true values. That is, the depth estimate is selected based on a tolerance chosen to reflect the desired probability that the estimate should deviate not more than a given percentage from the true value.

MRT letter: Recovering weak‐textured surfaces using image focus

In nature, objects have partially weak texture and their shape reconstruction using focus based passive methods like shape from focus (SFF), is difficult. This article presents a new SFF algorithm which can compute precise depth of dense as well as weak textured objects. Segmentation is applied to discard wrong depth estimate and then later interpolating them from accurate depth values of their neighbors. The performance of the proposed method is tested, using different image sequences of synthetic and real objects, with varying textures.