Eye tracking has already made its way to current commercial wearable display devices, and is becoming increasingly important for virtual and augmented reality applications. However, the existing model-based eye tracki...
详细信息
ISBN:
(数字)9781665453257
ISBN:
(纸本)9781665453257
Eye tracking has already made its way to current commercial wearable display devices, and is becoming increasingly important for virtual and augmented reality applications. However, the existing model-based eye tracking solutions are not capable of conducting very accurate gaze angle measurements, and may not be sufficient to solve challenging display problems such as pupil steering or eye-box expansion. In this paper, we argue that accurate detection and localization of pupil in 3D space is a necessary intermediate step in model-based eye tracking. Existing methods and datasets either ignore evaluating the accuracy of 3D pupil localization or evaluate it only on synthetic data. To this end, we capture the first 3D pupil-gaze-measurement dataset using a high precision setup with head stabilization and release it as the first benchmark dataset to evaluate both 3D pupil localization and gaze tracking methods. Furthermore, we utilize an advanced eye model to replace the commonly used oversimplified eye model. Leveraging the eye model, we propose a novel 3D pupil localization method with a deep learning-based corneal refraction correction. We demonstrate that our method outperforms the state-of-the-art works by reducing the 3D pupil localization error by 47.5% and the gaze estimation error by 18.7%. Our dataset and codes can be found here: link.
This paper presents Shape Aware Haptic Retargeting, an extension of "state-of-the-art" haptic retargeting that is the first to support retargeted interaction between any part of the user's hand and any p...
详细信息
ISBN:
(纸本)9781665496179
This paper presents Shape Aware Haptic Retargeting, an extension of "state-of-the-art" haptic retargeting that is the first to support retargeted interaction between any part of the user's hand and any part of the target object. In previous haptic retargeting algorithms, the maximum retargeting is applied only when the hand position aligns with the target position. Shape Aware Haptic Retargeting generalizes the distance computation process to instead consider the hand and target geometry. The shortest hand-target distance is then used to calculate the applied retargeting offset. This ensures the full amount of haptic retargeting is applied at the point of contact with the passive haptic regardless of contact position on the hand or target. We leverage existing geometry algorithms to implement three distance computation methods: Multi-Point, Primitive and Mesh Geometry, in addition to conventional single position approaches. These are evaluated through a set of simulated interactions instead of the single position representation used in previous haptic retargeting systems. The evaluation demonstrated all three approaches can provide improved interaction accuracy over a Point distance computation method, with Mesh Geometry being the most accurate and Primitive being the preferred method for combined performance and interaction accuracy.
Introduction/Background: The goal of Total Skin Electron Therapy (TSET) is to achieve a uniform surface dose, although assessment of this is never really done and typically limited points are sampled. A computational ...
详细信息
Introduction/Background: The goal of Total Skin Electron Therapy (TSET) is to achieve a uniform surface dose, although assessment of this is never really done and typically limited points are sampled. A computational treatment simulation approach was developed to estimate dose distributions over the body surface, to compare uniformity of (i) the 6 pose Stanford technique and (ii) the rotational technique. Methods: The relative angular dose distributions from electron beam irradiation was calculated by Monte Carlo simulation for cylinders with a range of diameters, approximating body part curvatures. These were used to project dose onto a 3D body model of the TSET patient's skin surfaces. computer animation methods were used to accumulate the dose values, for display and analysis of the homogeneity of cover-age. Results: The rotational technique provided more uniform coverage than the Stanford technique. Anomalies of under dose were observed in lateral abdominal regions, above the shoulders and in the perineum. The Stanford technique had larger areas of low dose laterally. In the rotational technique, 90% of the patient's skin was within +/- 10% of the prescribed dose, while this percentage decreased to 60% or 85% for the Stanford technique, varying with patient body mass. Interestingly, the highest discrepancy was most apparent in high body mass patients, which can be attributed to the loss of tangent dose at low angles of curvature. Discussion/Conclusion: This simulation and visualization approach is a practical means to analyze TSET dose, requiring only optical sur-face body topography scans. Under-and over-exposed body regions can be found, and irradiation could be customized to each patient. Dose Area Histogram (DAH) distribution analysis showed the rota-tional technique to have better uniformity, with most areas within 10% of the umbilicus value. Future use of this approach to analyze dose coverage is possible as a routine planning tool.
In minimally invasive surgeries one common way to verify progress is the use of an intraoperative X-ray device (due to its characteristic shape called a C-arm). Its control, however, remains challenging owing to its c...
详细信息
ISBN:
(纸本)9781665484022
In minimally invasive surgeries one common way to verify progress is the use of an intraoperative X-ray device (due to its characteristic shape called a C-arm). Its control, however, remains challenging owing to its complex movements. We propose the use of an Augmented Reality Head-Mounted Display (AR-HMD) to let the surgeon choose a desired X-ray view interventionally providing the corresponding C-arm configuration as visual feedback. The study participants' feedback, despite being critical of the HMD hardware limitations, suggests an inclination towards using AR for orthopaedic surgeries on especially complex or unusual anatomies.
Three-Dimensional (3D) object recognition has been receiving more and more attention in the field of computer Vision, as the development of AI algorithms and advancements of hardware components allow for better result...
详细信息
ISBN:
(纸本)9781665479332
Three-Dimensional (3D) object recognition has been receiving more and more attention in the field of computer Vision, as the development of AI algorithms and advancements of hardware components allow for better results nowadays more than ever before. Neural Networks are used to better classify different types of objects, these are being fed with pre-processed 3D objects represented in a 2D domain for better highlighting the important features and for a better performance. In this paper, we propose a strategy of feeding 2D representations of 3D meshes made by using different types of slicing to a CNN. In this way, we want to minimize the information loss and the redundancy the data suffers going from a 3D representation to a 2D one. To assess the results, we used Princeton's state-of-the-art ModelNet10 dataset, consisting of 4899 3D objects divided into 10 categories.
The Monte Carlo method, renowned for its ability to handle the spectral and geometric complexities of 3D radiative transfer, is extensively utilized across various fields, including concentrated solar power design, at...
详细信息
The Monte Carlo method, renowned for its ability to handle the spectral and geometric complexities of 3D radiative transfer, is extensively utilized across various fields, including concentrated solar power design, atmospheric science, and computer graphics. The success of this method also extends to the estimation of sensitivity-the derivative of an observable with respect to a given system parameter, which is, however, particularly challenging when these parameters involve geometric deformation. Bridging statistical physics and computer graphics, distinct methodologies have emerged within these fields for estimating geometric sensitivity, each employing unique terminologies and mathematical frameworks, leading to seemingly disparate approaches. In this paper, we review the three main approaches to sensitivity estimation: (1) Expectation Differentiation, which employs a vectorized Monte Carlo algorithm to simultaneously estimate the intensity and its sensitivity;(2) Differentiable Rendering, predominantly used in computer graphics and applied in numerous contexts;(3) Transport Model for Sensitivity, which conceptualizes sensitivity as a physical quantity with its own transport equations and boundary conditions, thereby facilitating engineering and physics analyses. We aim to enhance readers' ability to tackle sensitivity-related challenges by providing a comparative understanding of these three perspectives. We achieve this through a simplified one-dimensional radiative transfer case study, offering an accessible platform for comparing and classifying these approaches based on their theoretical underpinnings and practical application in Monte Carlo algorithms.
Transformative portals are a novel technique for supporting visualizations and interactions between locations that are not co-located in virtual reality (VR). We are interested in improving upon current portal renderi...
详细信息
ISBN:
(数字)9781665453257
ISBN:
(纸本)9781665453257
Transformative portals are a novel technique for supporting visualizations and interactions between locations that are not co-located in virtual reality (VR). We are interested in improving upon current portal rendering for VR, and supporting VR users to create new portals quickly and accurately. In this paper, we introduce a new high-level algorithm for rendering transformative portals in VR and present the benefits of our approach. By leveraging single-pass stereo and stencil portal rendering, we developed our algorithm to support multiple portals with "infinite" recursion, while running performant on mobile VR devices. We then benchmarked our performance against other common portal rendering implementations. In addition, we focused our research on VR interactions for handheld portals, in which the initial placement of the portal's destination is imperative to its effectiveness. In this paper, we present four new portal creation interactions for handheld portals: fishing reel, raycast with fishing reel, marker, and projectile curve. Based on existing research, we also established standard controls for fine-grained portal manipulation. In a user study, we compared and evaluated how well the creation interactions performed and argue that projectile curve is the most suitable general-purpose technique.
Light field displays represent yet another step in continually increasing pixel counts. Rendering realistic real-time 3D content for them with ray tracing-based methods is a major challenge even accounting for recent ...
详细信息
ISBN:
(纸本)9781450394659
Light field displays represent yet another step in continually increasing pixel counts. Rendering realistic real-time 3D content for them with ray tracing-based methods is a major challenge even accounting for recent hardware acceleration features, as renderers have to scale to tens to hundreds of distinct viewpoints. To this end, we contribute an open-source, cross-platform real-time 3D renderer called Tauray. The primary focus of Tauray is in using photorealistic path tracing techniques to generate real-time content for multi-view displays, such as VR headsets and light field displays;this aspect is generally overlooked in existing renderers. Ray tracing hardware acceleration aswell as multi-GPU rendering is supported. We compare Tauray to other open source real-time path tracers, like Lighthouse 2, and show that it can meet or significantly exceed their performance.
Projector-camera systems have long been used in measuring three-dimensional shapes. Most projector-camera systems can only be used in dark rooms because frame-based cameras are not robust against strong ambient light ...
详细信息
ISBN:
(纸本)9781665496179
Projector-camera systems have long been used in measuring three-dimensional shapes. Most projector-camera systems can only be used in dark rooms because frame-based cameras are not robust against strong ambient light and are difficult to obtain correspondence to the image pixels of the projector. Recently, event cameras, which can detect the direction of luminance change, have received attention in the field of computer vision. When considering the many advantages of event cameras, this study focuses on their wide dynamic range (120 vs. 40 dB of a frame-based camera) and their ability to detect fast luminance changes. Our objective is to realize a projector-camera system that combines the event camera with a projector under the strong ambient light. Specifically, this study proposes a new structured light that combines different frequencies of flickers to acquire the correspondence between the image pixels of the event camera and the projector. This method does not rely on the co-axial frame-based measurement and synchronization mechanism between projector and camera and is thus applicable to most general event cameras. Experiments confirm that the proposed method obtains the correspondence robustly with reasonable accuracy in a bright room (up to 2,600 lux) under general indoor lighting and additional light projection.
Physics-based rendering (PBR) engines attempt to generate photorealistic images by mimicking light-matter interaction in a physically plausible way. PBR has become the standard rendering method in the fields of animat...
详细信息
ISBN:
(纸本)9781510651012;9781510651005
Physics-based rendering (PBR) engines attempt to generate photorealistic images by mimicking light-matter interaction in a physically plausible way. PBR has become the standard rendering method in the fields of animation, gaming, and computer graphics research. More recently, PBR engines have included the ability to track the full polarization state of light. An area of interest for polarization-aware PBR engines is validating the accuracy of polarized bi-direction reflection distribution functions (pBRDF). pBRDFs are polarized material models described by a geometry-, texture-, and albedo-dependent Mueller matrix. For renderings, methods to analyze the pBRDF are limited. This work presents a pBRDF analysis method that simulates a Mueller matrix imaging polarimeter using a polarization-aware PBR engine. Simulated reconstructed Mueller matrix images are qualitatively compared to measurements from a Mueller matrix imaging polarimeter.
暂无评论