Virtual Shadows Research Articles

Data Augmentation for Semantic Segmentation Using a Real Image Dataset Captured Around the Tsukuba City Hall

We are exploring the use of semantic scene understanding in autonomous navigation for the Tsukuba Challenge. However, manually creating a comprehensive dataset that covers various outdoor scenes with time and weather variations to ensure high accuracy in semantic segmentation is onerous. Therefore, we propose modifications to the model and backbone of semantic segmentation, along with data augmentation techniques. The data augmentation techniques, including the addition of virtual shadows, histogram matching, and style transformations, aim to improve the representation of variations in shadow presence and color tones. In our evaluation using images from the Tsukuba Challenge course, we achieved the highest accuracy by switching the model to PSPNet and changing the backbone to ResNeXt. Furthermore, the adaptation of shadow and histogram proved effective for critical classes in robot navigation, such as road, sidewalk, and terrain. In particular, the combination of histogram matching and shadow application demonstrated effectiveness for data not included in the base training dataset.

Shadow induction on optical see-through head-mounted displays

In this paper, we present two effects of illusion-based virtual shadows on a real surface for optical see-through head-mounted displays without devices that selectively reduce or intercept light rays from the real scene. The virtual shadows are based on the simultaneous brightness-contrast illusion in which a central region surrounded by a lighter one is perceived as darker than one without such a lighter region. To display a virtual shadow, a virtual object with a real image texture is placed around the shadow region as the lighter region to provide a sense of a darker area. We conducted two kinds of experiments to validate the feasibility of our method. The first experiment showed that users tend to perceive the virtual shadow part as darker than the surrounding area with the same physical luminance. The second experiment provided evidence that there are cases in which the virtual shadows can improve the reality of virtual objects.

Influence of virtual objects' shadows and lighting coherence on distance perception in optical see‐through augmented reality

AbstractThis paper focuses on how virtual objects' shadows as well as differences in alignment between virtual and real lighting influence distance perception in optical see‐through (OST) augmented reality (AR). Four hypotheses are proposed: (H1) Participants underestimate distances in OST AR; (H2) Virtual objects' shadows improve distance judgment accuracy in OST AR; (H3) Shadows with different realism levels have different influence on distance perception in OST AR; (H4) Different levels of lighting misalignment between real and virtual lights have different influence on distance perception in OST AR scenes. Two experiments were designed with an OST head mounted display (HMD), the Microsoft HoloLens. Participants had to match the position of a virtual object displayed in the OST‐HMD with a real target. Distance judgment accuracy was recorded under the different shadows and lighting conditions. The results validate hypotheses H2 and H4 but surprisingly showed no impact of the shape of virtual shadows on distance judgment accuracy thus rejecting hypothesis H3. Regarding hypothesis H1, we detected a trend toward underestimation; given the high variance of the data, more experiments are needed to confirm this result. Moreover, the study also reveals that perceived distance errors and completion time of trials increase along with targets' distance.

Enhanced Shadow Retargeting with Light-Source Estimation Using Flat Fresnel Lenses

Shadow-retargeting maps depict the appearance of real shadows to virtual shadows given corresponding deformation of scene geometry, such that appearance is seamlessly maintained. By performing virtual shadow reconstruction from unoccluded real-shadow samples observed in the camera frame, this method efficiently recovers deformed shadow appearance. In this manuscript, we introduce a light-estimation approach that enables light-source detection using flat Fresnel lenses that allow this method to work without a set of pre-established conditions. We extend the adeptness of this approach by handling scenarios with multiple receiver surfaces and a non-grounded occluder with high accuracy. Results are presented on a range of objects, deformations, and illumination conditions in real-time Augmented Reality (AR) on a mobile device. We demonstrate the practical application of the method in generating otherwise laborious in-betweening frames for 3D printed stop-motion animation.

LivePhantom: Retrieving Virtual World Light Data to Real Environments.

To achieve realistic Augmented Reality (AR), shadows play an important role in creating a 3D impression of a scene. Casting virtual shadows on real and virtual objects is one of the topics of research being conducted in this area. In this paper, we propose a new method for creating complex AR indoor scenes using real time depth detection to exert virtual shadows on virtual and real environments. A Kinect camera was used to produce a depth map for the physical scene mixing into a single real-time transparent tacit surface. Once this is created, the camera’s position can be tracked from the reconstructed 3D scene. Real objects are represented by virtual object phantoms in the AR scene enabling users holding a webcam and a standard Kinect camera to capture and reconstruct environments simultaneously. The tracking capability of the algorithm is shown and the findings are assessed drawing upon qualitative and quantitative methods making comparisons with previous AR phantom generation applications. The results demonstrate the robustness of the technique for realistic indoor rendering in AR systems.

Virtual Shadow

Most obvious benefit of augmented reality (AR) displays is direct perception of information atop physical reality. In driving context, however, AR interfaces should be designed carefully to guide drivers’ attention while minimizing attentional narrowing. This work aims to design an interface for cross traffic alert using an AR head up display (HUD) that is compatible with both the driver’s cognitive process and physical reality of driving environment. Ecological interface design (EID) allowed us to complement current user centered design (UCD) approaches by considering human-environment interaction and leveraging the inherent benefit of AR interfaces: conformal graphics. We designed a novel interface that casts virtual shadows of approaching obstacles through an AR HUD and prototyped this idea for a specific use-case of pedestrian collision warning. Our initial usability evaluation demonstrated potential benefits of incorporating EID into AR interface design. The approaches and design idea from this study can be leveraged by future researchers and designers to create more reliable and safer AR interfaces for vehicle drivers.

Under Surveillance

Karen Lawrence Öqvist is the author of the forthcoming BCS book, Virtual Shadows: Your Privacy in the Information Society, which is due out in December. Henry Tucker spoke to her.

II. On the sensitive state of vacuum discharges. Part II

This paper forms a sequel to that published under the same title in the “Phil. Trans.,” 1879, p. 165. It describes a continuation of the research into the nature and laws of the disruptive discharge, or electrie spark. The methods of the earlier paper have been extended, and others adapted to the new circumstances have been devised m order to carry the investigation into high vacua. In particular, independent sources of electricity have been used for affecting the discharge, whether in the sensitive or in the non-sensitive state; and the results have been confirmatory of the conclusions derived from the more limited means formerly described. Further, the effects of various tubes containing discharges in the sensitive state upon a tube containing a discharge in the non-sensitive state have been observed and compared; and the tube so used as a test has been called the standard tube, and the method of its use the standard tube method. By this means, principally, the laws of the discharge in comparatively moderate vacua have been extended to high vacua. In the higher vacua, the phenomena of molecular streams, and the phosphorescence consequent on them, that have been studied and described by Mr. Crookes, present themselves. These derive great importance for the purposes of the present paper from the fact that in high vacua the ordinary luminous discharge becomes so feeble m appearance that it is often difficult to observe. Under these circumstances the phosphorescence, which like the ordinary luminous effects may exist either in a sensitive or in a non-sensitive state, forms the best index of what is going on within the tube. Much information as to the nature and procedure of the discharge may be derived from the mode of interference of one molecular stream with another, from the direction and character of shadows cast by these streams, and by a form of interference which has here been called that of virtual shadows.