GeoNet++: Iterative Geometric Neural Network with Edge-Aware Refinement for Joint Depth and Surface Normal Estimation.

Photometric stereo is a method to obtain surface normals of an object using its images captured under varying illumination directions. The existing deep learning-based methods require multiple images of an object captured using complex image acquisition systems. In this work, we propose a deep learning framework to perform three tasks jointly: (i) lighting estimation, (ii) image relighting, and (iii) surface normal estimation, all from a single input image of an object with non-Lambertian surface and general reflectance. The network explicitly segregates global geometric features and local lighting-specific features of the object from a single image. The local features resemble attached shadows, shadings, and specular highlights, providing valuable lighting estimation and relighting cues. The global features capture the lighting-independent geometric attributes that effectively guide the surface normal estimation. The joint training transfers valuable insights to achieve significant improvements across all three tasks. We show that the proposed single-image-based relighting framework outperforms several existing photometric stereo methods which require multiple images of a static object.

<p><strong>This thesis explores innovative methods for enhancing the manipulation and geometric understanding of 360° images in virtual reality (VR) applications, with a focus on depth estimation, surface normal prediction, and multi-task learning. The research presents a comprehensive approach to improving the manipulation of stereo 360° images, particularly in the context of image composition. A novel method is proposed to seamlessly integrate new visual elements into stereo 360° scenes, thereby enhancing user interaction and immersion in VR environments.</strong></p><p>To support such advanced editing applications, a new technique for estimating depth maps from monocular 360° images is introduced. This method leverages both local and global scene information, significantly improving the accuracy of depth predictions. This is particularly crucial for applications that require spatial awareness, such as virtual and augmented reality, where accurate depth perception is key to realistic interaction with the virtual environment.</p><p>Furthermore, the thesis introduces a hybrid approach that combines convolutional neural networks (CNNs) and Vision Transformers (ViTs) to improve surface normal estimation. This method takes advantage of CNNs' ability to capture fine-grained details and ViTs' strength in modeling global context, resulting in more precise surface geometry analysis from 360° imagery. This enhanced surface normal estimation plays a vital role in better understanding the spatial structure of the scene.</p><p>In addition, the research demonstrates the effectiveness of multi-task learning (MTL) for comprehensive scene geometry understanding. By predicting both depth and surface normals simultaneously from monocular 360° images, the proposed MTL framework delivers a more detailed and coherent geometric representation, further contributing to the realism and immersion of VR used scenarios.</p><p>Through these contributions, the thesis significantly advances the field of 360° image-based manipulations and scene understanding, offering new tools and methodologies that pave the way for more immersive, interactive, and spatially aware experiences for VR applications.</p>

In the past few years, convolutional neural nets (CNN) have shown incredible promise for learning visual representations. In this paper, we use CNNs for the task of predicting surface normals from a single image. But what is the right architecture we should use? We propose to build upon the decades of hard work in 3D scene understanding, to design new CNN architecture for the task of surface normal estimation. We show by incorporating several constraints (man-made, manhattan world) and meaningful intermediate representations (room layout, edge labels) in the architecture leads to state of the art performance on surface normal estimation. We also show that our network is quite robust and show state of the art results on other datasets as well without any fine-tuning.

Deep-learning-based polarization 3D imaging techniques, which train networks in a data-driven manner, are capable of estimating a target’s surface normal distribution under passive lighting conditions. However, existing methods have limitations in restoring target texture details and accurately estimating surface normals. Information loss can occur in the fine-textured areas of the target during the reconstruction process, which can result in inaccurate normal estimation and reduce the overall reconstruction accuracy. The proposed method enables extraction of more comprehensive information, mitigates the loss of texture information during object reconstruction, enhances the accuracy of surface normal estimation, and facilitates more comprehensive and precise reconstruction of objects. The proposed networks optimize the polarization representation input by utilizing the Stokes-vector-based parameter, in addition to separated specular and diffuse reflection components. This approach reduces the impact of background noise, extracts more relevant polarization features of the target, and provides more accurate cues for restoration of surface normals. Experiments are performed using both the DeepSfP dataset and newly collected data. The results show that the proposed model can provide more accurate surface normal estimates. Compared to the UNet architecture-based method, the mean angular error is reduced by 19%, calculation time is reduced by 62%, and the model size is reduced by 11%.

Monocular depth estimation with guidance of surface normal map

This paper presents a novel method for 3D surface reconstruction that uses polarization and shading information from two views. The method relies on polarization data acquired using a standard digital camera and a linear polarizer. Fresnel theory is used to process the raw images and to obtain initial estimates of surface normals, assuming that the reflection type is diffuse. Based on this idea, the paper presents two novel contributions to the problem of surface reconstruction. The first is a technique to enhance the surface normal estimates by incorporating shading information into the method. This is done using robust statistics to estimate how the measured pixel brightnesses depend on the surface orientation. This gives an estimate of the object material reflectance function, which is used to refine the estimates of the surface normals. The second contribution is to use the refined estimates to establish correspondence between two views of an object. To do this, a set of patches are extracted from each view and are aligned by minimizing an energy functional based on the surface normal estimates and local topographic properties. The optimum alignment parameters for different patch pairs are then used to establish stereo correspondence. This process results in an unambiguous field of surface normals, which can be integrated to recover the surface depth. Our technique is most suited to smooth, non-metallic surfaces. It complements existing stereo algorithms since it does not require salient surface features to obtain correspondences. An extensive set of experiments, yielding reconstructed objects and reflectance functions, are presented and compared to ground truth.

We address the problem of reconstructing 3D surfaces from depth and surface normal maps acquired by a sensor system based on a single perspective camera. Depth and normal maps can be obtained through techniques such as structured‐light scanning and photometric stereo, respectively. We propose a perspective‐aware log‐depth fusion approach that extends existing orthographic gradient‐based depth‐normals fusion methods by explicitly accounting for perspective projection, leading to metrically accurate 3D reconstructions. Additionally, the method handles missing depth measurements by leveraging available surface normal information to inpaint gaps. Experiments on the DiLiGenT‐MV data set demonstrate the effectiveness of our approach and highlight the importance of perspective‐aware depth‐normals fusion.

We present a single-capture photometric stereo method using a hyperspectral camera. A spectrally and spatially designed illumination enables a point-wise estimation of reflectance spectra and surface normals from a single hyperspectral image. The illumination works as a reflectance probe in wide spectral regions where reflectance spectra are measured, and the full spectra are estimated by interpolation. It also works as the resource for shadings in other spectral regions. The accuracy of estimation is evaluated in a simulation. Also, we prepare an experimental setup and demonstrate a surface reconstruction against a real scene.

As part of the DOE Heliostat Consortium Roadmap study, we investigated optical metrology systems for heliostats, seeking areas where further research was needed. We began by considering optical metrology questions of interest across the heliostat development cycle and identified information types common across this spectrum. In addition to raising questions of interest, each development cycle phase implied specific operation requirements. Combining these yielded 13 core problem statements, four of which do not appear to have a readily available solution: (a) in-situ measurement of high-resolution maps of mirror surface normals, for all heliostat tilt angles and heliostats far from the tower; (b) accelerated heliostat calibration; (c) high-speed in situ measurement of heliostat surface normal maps and pointing directions; (d) ground truth methods for verifying the accuracy of surface normal map measurements. This analysis may provide input to the selection of future research goals.

In this paper, we introduce a deep multi-view stereo (MVS) system that jointly predicts depths, surface normals and per-view confidence maps. The key to our approach is a novel solver that iteratively solves for per-view depth map and normal map by optimizing an energy potential based on the locally planar assumption. Specifically, the algorithm updates depth map by propagating from neighboring pixels with slanted planes, and updates normal map with local probabilistic plane fitting. Both two steps are monitored by a customized confidence map. This solver is not only effective as a post-processing tool for plane-based depth refinement and completion, but also differentiable such that it can be efficiently integrated into deep learning pipelines. Our multi-view stereo system employs multiple optimization steps of the solver over the initial prediction of depths and surface normals. The whole system can be trained end-to-end, decoupling the challenging problem of matching pixels within poorly textured regions from the cost-volume based neural network. Experimental results on ScanNet and RGB-D Scenes V2 demonstrate state-of-the-art performance of the proposed deep MVS system on multi-view depth estimation, with our proposed solver consistently improving the depth quality over both conventional and deep learning based MVS pipelines. Code is available at https://github.com/thuzhaowang/idn-solver.

We introduce an approach that leverages surface normal predictions, along with appearance cues, to retrieve 3D models for objects depicted in 2D still images from a large CAD object library. Critical to the success of our approach is the ability to recover accurate surface normals for objects in the depicted scene. We introduce a skip-network model built on the pre-trained Oxford VGG convolutional neural network (CNN) for surface normal prediction. Our model achieves state-of-the-art accuracy on the NYUv2 RGB-D dataset for surface normal prediction, and recovers fine object detail compared to previous methods. Furthermore, we develop a two-stream network over the input image and predicted surface normals that jointly learns pose and style for CAD model retrieval. When using the predicted surface normals, our two-stream network matches prior work using surface normals computed from RGB-D images on the task of pose prediction, and achieves state of the art when using RGB-D input. Finally, our two-stream network allows us to retrieve CAD models that better match the style and pose of a depicted object compared with baseline approaches.

Photometric stereo (PST) is a widely used technique of estimating surface normals from an image set. However, it often produces inaccurate results for non-Lambertian surface reflectance. In this study, PST is reformulated as a sparse recovery problem where non-Lambertian errors are explicitly identified and corrected. We show that such a problem can be accurately solved via a greedy algorithm called orthogonal matching pursuit (OMP). The performance of OMP is evaluated on synthesized and real-world datasets: we found that the greedy algorithm is overall more robust to non-Lambertian errors than other state-of-the-art sparse approaches with little loss of efficiency. Along with providing an overview of current methods, novel contributions in this paper are as follows: we propose an alternative sparse formulation for PST; in previous PST studies (Wu et al., Robust photometric stereo via low-rank matrix completion and recovery, 2010), (S. Ikehata et al., Robust photometric stereo using sparse regression, 2012), the surface normal vector and the error vector are treated as two entities and are solved independently. In this study, we convert their formulation into a new canonical form of the sparse recovery problem by combining the two vectors into one large vector in a new “stacked” formulation in this domain. This allows for a large repertoire of existing sparse recovery algorithms to be more straightforwardly applied to the PST problem. In our application of the OMP greedy algorithm, we show that greedy solvers can indeed be applied, with this study supplying the first of such attempt at employing greedy approaches to estimate surface normals within the framework of PST. We numerically compare the performance of several normal vector recovery methods. Most notably, this is the first detailed test on complex images of the normal estimation accuracy of our previously proposed method, least median of squares (LMS).

This paper presents symmetric-light photometric stereo for surface normal estimation, in which directional lights are distributed symmetrically with respect to the optic center. Unlike previous studies of ring-light settings that required the information of ring radius, we show that even without the knowledge of the exact light source locations or their distances from the optic center, the symmetric configuration provides us sufficient information for recovering unique surface normals without ambiguity. Specifically, under the symmetric lights, measurements of a pair of scene points having distinct surface normals but the same albedo yield a system of constrained quadratic equations about the surface normal, which has a unique solution. Experiments demonstrate that the proposed method alleviates the need for geometric light source calibration while maintaining the accuracy of calibrated photometric stereo.

Estimating depth information from a single image has recently attracted great attention in various vision-based applications such as mobile robot navigation. Although there are numerous depth map generation methods, little effort has been done on the depth estimation from a single indoor scene. In this paper, we propose a novel method for estimating depth from a single indoor image via nonlinear diffusion and image segmentation techniques. One important advantage of our approach is that no learning scheme is required to estimate a depth map. Based on the proposed method, we obtain visually plausible depth estimation results even with the presence of occlusions or clutters in the single indoor image. From experimental results, we confirm that the proposed algorithm provides reliable depth information under various indoor environments.

Depth estimation and camera calibration of a focused plenoptic camera for visual odometry