Aesthetic experiences are closely related to human emotions and experiences as well as the perception of aesthetic objects and phenomena. We consider interaction with generated immersive worlds to be an immersive aest...
详细信息
ISBN:
(数字)9798331514846
ISBN:
(纸本)9798331525637
Aesthetic experiences are closely related to human emotions and experiences as well as the perception of aesthetic objects and phenomena. We consider interaction with generated immersive worlds to be an immersive aesthetic experience if it is designed independently of any functional use and can be experienced in an aesthetically pleasing way. The possibilities of generative techniques such as AI and other algorithms and visualization through the integration of virtual reality open up new aesthetic possibilities. We present the results of a collaboration between scientists and designers/artists in the development of a framework for three-dimensional fractals [5] and its application in various scenarios in the field of festivals, performances and exhibitions.
This paper presents new hierarchically cascaded transformers that can improve data efficiency through attribute surrogates learning and spectral tokens pooling. Vision transformers have recently been thought of as a p...
详细信息
ISBN:
(数字)9781665469463
ISBN:
(纸本)9781665469463
This paper presents new hierarchically cascaded transformers that can improve data efficiency through attribute surrogates learning and spectral tokens pooling. Vision transformers have recently been thought of as a promising alternative to convolutional neural networks for visual recognition. But when there is no sufficient data, it gets stuck in overfitting and shows inferior performance. To improve data efficiency, we propose hierarchically cascaded transformers that exploit intrinsic image structures through spectral tokens pooling and optimize the learnable parameters through latent attribute surrogates. The intrinsic image structure is utilized to reduce the ambiguity between foreground content and background noise by spectral tokens pooling. And the attribute surrogate learning scheme is designed to benefit from the rich visual information in image-label pairs instead of simple visual concepts assigned by their labels. Our Hierarchically Cascaded Transformers, called HCTransformers, is built upon a self-supervised learning framework DINO and is tested on several popular few-shot learning benchmarks. In the inductive setting, HCTransformers surpass the DINO baseline by a large margin of 9.7% 5-way 1-shot accuracy and 9.17% 5-way 5-shot accuracy on miniImageNet, which demonstrates HCTransformers are efficient to extract discriminative features. Also, HCTransformers show clear advantages over SOTA few-shot classification methods in both 5-way 1-shot and 5-way 5-shot settings on four popular benchmark datasets, including miniImageNet, tieredImageNet, FC100, and CIFAR-FS. The trained weights and codes are available at https://***/StomachCold/HCTransformers.
Methods of presenting visual information, known as phosphenes, to visually impaired people have been researched. Phosphenes are perceived by applying electrical stimulation to the visual pathway, and the position of t...
详细信息
ISBN:
(数字)9798350340204
ISBN:
(纸本)9798350340211
Methods of presenting visual information, known as phosphenes, to visually impaired people have been researched. Phosphenes are perceived by applying electrical stimulation to the visual pathway, and the position of the electrodes can control the presentation position. However, presenting one phosphenes requires two electrodes, and presenting multiple phosphenes requires placing many electrodes on the face. Therefore, this study discusses a method that reduces the number of electrodes placed in the limited space on the human face and presents multiple phosphenes.
The pharmaceutical industry facing challenges despite digital transformation applying for the benefit of the industry. The companies and the industrial process need to be ready and adapt to the rapid changes or else b...
详细信息
Process monitoring plays a major role in ensuring the safety and efficiency of industrial processes. This paper introduces a new concept of normal operating zones for process monitoring. The normal operating zone (NOZ...
详细信息
ISBN:
(数字)9798331521950
ISBN:
(纸本)9798331521967
Process monitoring plays a major role in ensuring the safety and efficiency of industrial processes. This paper introduces a new concept of normal operating zones for process monitoring. The normal operating zone (NOZ) is a high-dimensional geometric space consisting of variation ranges of multiple related process variables in normal conditions. It provides spatial information among process variables, as an indispensable complement to time information provided by time series of process variables. Abnormality detection and operation guidance for improved efficiency can be realized for multivariate processes by integrating time and spatial information in a visual and transparent way. Experimental and industrial case studies are provided to illustrate the applications of NOZs on threetank systems and power generation units.
Photogrammetry is one of the non-invasive markerless methods applied in postural assessment to measure the dynamical relation of body parts in terms of distance and angles. The use of camera sensing technology replace...
详细信息
ISBN:
(纸本)9781665486637
Photogrammetry is one of the non-invasive markerless methods applied in postural assessment to measure the dynamical relation of body parts in terms of distance and angles. The use of camera sensing technology replaces visual inspectors to quantify the postural assessment in a more efficient way. To visualize a series of human motion, the proposed web-app automates the process of photogrammetry and generates the kinematic analysis. It allows three-dimensional joints visualization and an interactive kinematic graph for joints correlation. A preliminary postural assessment can be completed via uploading a video or a text file with postural information. The output of the web-app can be used to evaluate the posture of the human from the input data using 3D human joints, distance, and joint movement. In this work, the walking and running patterns were put into practise and examined. Results show that the kinematic analysis is made possible by the visualisation of 3D human joints in euclidean distance and joint angular characteristics.
Estimating precise metric depth and scene reconstruction from monocular endoscopy is a fundamental task for surgical navigation in robotic surgery. However, traditional stereo matching adopts binocular images to perce...
详细信息
ISBN:
(数字)9781665479271
ISBN:
(纸本)9781665479271
Estimating precise metric depth and scene reconstruction from monocular endoscopy is a fundamental task for surgical navigation in robotic surgery. However, traditional stereo matching adopts binocular images to perceive the depth information, which is difficult to transfer to the soft robotics-based surgical systems due to the use of monocular endoscopy. In this paper, we present a novel framework that combines robot kinematics and monocular endoscope images with deep unsupervised learning into a single network for metric depth estimation and then achieve 3D reconstruction of complex anatomy. Specifically, we first obtain the relative depth maps of surgical scenes by leveraging a brightness-aware monocular depth estimation method. Then, the corresponding endoscope poses are computed based on non-linear optimization of geometric and photometric reprojection residuals. Afterwards, we develop a Depth-driven Sliding Optimization (DDSO) algorithm to extract the scaling coefficient from kinematics and calculated poses offline. By coupling the metric scale and relative depth data, we form a robust ensemble that represents the metric and consistent depth. Next, we treat the ensemble as supervisory labels to train a metric depth estimation network for surgeries (i.e., MetricDepthS-Net) that distills the embeddings from the robot kinematics, endoscopic videos, and poses. With accurate metric depth estimation, we utilize a dense visual reconstruction method to recover the 3D structure of the whole surgical site. We have extensively evaluated the proposed framework on public SCARED and achieved comparable performance with stereo-based depth estimation methods. Our results demonstrate the feasibility of the proposed approach to recover the metric depth and 3D structure with monocular inputs.
As patient numbers rise, managing antibiotic usage and adhering to Antimicrobial Stewardship Program (ASP) guidelines becomes increasingly challenging for healthcare professionals. This study integrates Power BI with ...
详细信息
Narrative visualization has become a crucial tool in data presentation, merging storytelling with data visualization to convey complex information in an engaging and accessible manner. In this study, we review the des...
详细信息
ISBN:
(数字)9798350354850
ISBN:
(纸本)9798350354867
Narrative visualization has become a crucial tool in data presentation, merging storytelling with data visualization to convey complex information in an engaging and accessible manner. In this study, we review the design space for narrative visualizations, focusing on animation style, through a comprehensive analysis of 80 papers from key visualization venues. We categorize these papers into six broad themes: Animation Style, Interactivity, Technology Usage, Methodology Development, Evaluation Type, and Application Domain. Our findings reveal a significant evolution in the field, marked by a growing preference for animated and non-interactive techniques. This trend reflects a shift towards minimizing user interaction while enhancing the clarity and impact of data presentation. We also identified key trends and technologies shaping the field, highlighting the role of technologies, such as machine learning in driving these changes. We offer insights into the dynamic interrelations within the narrative visualization domains, and suggest future research directions, including exploring non-interactive techniques, examining the interplay between different visualization elements, and developing domain-specific visualizations.
暂无评论