Recently unsupervised learning of depth from videos has made remarkable progress and the results are comparable to fully supervised methods in outdoor scenes like KITTI. However, there still exist great challenges when directly applying this technology in indoor environments, e.g., large areas of non-texture regions like white wall, more complex ego-motion of handheld camera, transparent glasses and shiny objects. To overcome these problems, we propose a new optical-flow based training paradigm which reduces the difficulty of unsupervised learning by providing a clearer training target and handles the non-texture regions. Our experimental evaluation demonstrates that the result of our method is comparable to fully supervised methods on the NYU Depth V2 benchmark. To the best of our knowledge, this is the first quantitative result of purely unsupervised learning method reported on indoor datasets.
A new efficient biorthogonal wavelet analysis based on the principal square root of subdivision is proposed in the paper by using the lifting scheme. Since the principal square root of subdivision is of the slowest topological refinement among the traditional triangular subdivisions, the multiresolution analysis based on the principal square root of subdivision is more balanced than the existing wavelet analyses on triangular meshes, and accordingly offers more levels of detail for processing polygonal models. In order to optimize the multiresolution analysis process, the new wavelets, no matter whether they are interior or on boundaries, are orthogonalized with the local scaling functions based on a discrete inner product with subdivision masks. Because the wavelet analysis and synthesis algorithms are actually composed of a series of local lifting operations, they can be performed in linear time. The experiments demonstrate the efficiency and stability of the wavelet analysis for both closed and open triangular meshes with principal square root of subdivision connectivity. The principal square root of -subdivision-based biorthogonal wavelets can be used in many applications such as progressive transmission, shape approximation, multiresolution editing and rendering of 3D geometric models.
Unsupervised depth learning takes the appearance difference between a target view and a view synthesized from its adjacent frame as supervisory signal. Since the supervisory signal only comes from images themselves, the resolution of training data significantly impacts the performance. High-resolution images contain more fine-grained details and provide more accurate supervisory signal. However, due to the limitation of memory and computation power, the original images are typically down-sampled during training, which suffers heavy loss of details and disparity accuracy. In order to fully explore the information contained in high-resolution data, we propose a simple yet effective dual networks architecture, which can directly take highresolution images as input and generate high-resolution and high-accuracy depth map efficiently. We also propose a Self-assembled Attention (SA-Attention) module to handle low-texture region. The evaluation on the benchmark KITTI and Make3D datasets demonstrates that our method achieves state-of-the-art results in the monocular depth estimation task.
Reflections, refractions, and caustics are very important for rendering global illumination images. Although many methods can be applied to generate these effects, the rendering performance is not satisfactory for interactive applications. In this paper, complex ray-object intersections are simplified so that the intersections can be computed on a GPU, and an iterative computing scheme based on the depth buffers is used for correcting the approximate results caused by the simplification. As a result, reflections and refractions of environment maps and nearby geometry can be rendered on a GPU interactively without preprocessing. We can even achieve interactive recursive reflections and refractions by using an object-impostor technique. Moreover, caustic effects caused by reflections and refractions can be rendered by placing the eye at the light. Rendered results prove that our method is sufficiently efficient to render plausible images interactively for many interactive applications.
In ultrasonic nondestructive testing (NDT), the phase shift migration (PSM) technique, as a frequency-domain implementation of the synthetic aperture focusing technique (SAFT), can be adopted for imaging of regularly layered objects that are inhomogeneous only in depth but isotropic and homogeneous in the lateral direction. To deal with irregularly layered objects that are anisotropic and inhomogeneous in both the depth and lateral directions, a generalized frequency- domain SAFT, called generalized phase shift migration (GPSM), is proposed in this paper. Compared with PSM, the most significant innovation of GPSM is that the phase shift factor is generalized to handle anisotropic media with lateral velocity variations. The generalization is accomplished by computer programming techniques without modifying the PSM model. In addition, SRFFT (split-radix fast Fourier transform) input/output pruning algorithms are developed and employed in the GPSM algorithm to speed up the image reconstructions. The experiments show that the proposed imaging techniques are capable of reconstructing accurate shapes and interfaces of irregularly layered objects. The computing time of the GPSM algorithm is much less than the time-domain SAFT combined with the ray-tracing technique, which is, at present, the common method used in ultrasonic NDT industry for imaging layered objects. Furthermore, imaging regularly layered objects can be regarded as a special case of the presented technique.
In this paper, we propose and analyze a subdivision scheme which unifies 3-point approximating subdivision schemes of any arity in its compact form and has less support, computational cost and error bounds. The usefulness of the scheme is illustrated by considering different examples along with its comparison with the established subdivision schemes. Moreover, B-splines of degree 4and well known 3-point schemes [1, 2, 3, 4, 6, 11, 12, 14, 15] are special cases of our proposed scheme.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.