Related papers: Stereoscopic visualization in curved spacetime: se…
The distortion of the spacetime structure in the surroundings of black holes affects the trajectories of light rays. As a consequence, black holes can act as gravitational lenses. Observations of type Ia supernovas, show that our Universe…
Monocular depth prediction plays a crucial role in understanding 3D scene geometry. Although recent methods have achieved impressive progress in terms of evaluation metrics such as the pixel-wise relative error, most methods neglect the…
Learning a latent embedding to understand the underlying nature of data distribution is often formulated in Euclidean spaces with zero curvature. However, the success of the geometry constraints, posed in the embedding space, indicates that…
Human visual system relies on both binocular stereo cues and monocular focusness cues to gain effective 3D perception. In computer vision, the two problems are traditionally solved in separate tracks. In this paper, we present a unified…
We introduce algorithms to visualize feature spaces used by object detectors. The tools in this paper allow a human to put on `HOG goggles' and perceive the visual world as a HOG based object detector sees it. We found that these…
Since the development of Relativity theory, a variety of solutions have been found around Einstein field equations, depending on the time-space. For instance, we can find the axially symmetric space which describes a rotanting black hole.…
Can the observation of the "shadow" allow us to distinguish a black hole from a more exotic compact object? We study the motion of photons in a class of vacuum static axially-symmetric space-times that is continuously linked to the…
Special relativity turns out to be more than coordinate transformations in which the constancy of the speed of light plays the central role between two inertial reference frames. Special relativity, in essence, is a theory of…
We introduce algorithms to visualize feature spaces used by object detectors. Our method works by inverting a visual feature back to multiple natural images. We found that these visualizations allow us to analyze object detection systems in…
Neural approaches have shown a significant progress on camera-based reconstruction. But they require either a fairly dense sampling of the viewing sphere, or pre-training on an existing dataset, thereby limiting their generalizability. In…
Stereoscopic 3D (S3D) displays provide an additional sense of depth compared to non-stereoscopic displays by sending slightly different images to the two eyes. But conventional S3D displays do not reproduce all natural depth cues. In…
Stereo vision is an effective technique for depth estimation with broad applicability in autonomous urban and highway driving. While various deep learning-based approaches have been developed for stereo, the input data from a binocular…
Stereophotogrammetry is an established technique for scene understanding. Its origins go back to at least the 1800s when people first started to investigate using photographs to measure the physical properties of the world. Since then,…
Stereo matching is one of the longest-standing problems in computer vision with close to 40 years of studies and research. Throughout the years the paradigm has shifted from local, pixel-level decision to various forms of discrete and…
The impact of curvature divergences on physical observers in a black hole space-time which, nonetheless, is geodesically complete is investigated. This space-time is an exact solution of certain extensions of General Relativity coupled to…
Even in free space, which is commonly considered of as a flat space-time in most settings, the degree of polarization of a partially spatially coherent light beam changes as it travels. Similarly, the polarization degree would change when a…
First sketch of black hole from M87 galaxy was obtained by Event Horizon Telescope, recently. As appearance of black hole shadow reflects space-time geometry of black holes, observations of black hole shadow may be a promising way to test…
Stereoscopic head-mounted displays (HMDs) render and present binocular images to create an egocentric, 3D percept to the HMD user. Within this render and presentation pipeline there are potential rendering camera and viewing position errors…
Recently, end-to-end trainable deep neural networks have significantly improved stereo depth estimation for perspective images. However, 360{\deg} images captured under equirectangular projection cannot benefit from directly adopting…
The measurement of distance between two objects is generalized to the case where the objects are no longer points but are one-dimensional. Additional concepts such as non-extensibility, curvature constraints, and non-crossing become central…