Article
Computer Science, Software Engineering
Budmonde Duinkharjav, Kenneth Chen, Abhishek Tyagi, Jiayi He, Yuhao Zhu, Qi Sun
Summary: Battery life is a pressing issue for wireless VR and AR devices. Researchers have developed a gaze-contingent system that reduces display power consumption without compromising visual fidelity, addressing this challenge.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Software Engineering
Hui Zeng, Rong Zhao
Summary: We propose a dual-mode virtual reality system based on the spatio-temporal perception characteristics of human vision. This system can adaptively adjust the display spatial and temporal resolution to provide users with optimal visual perception quality.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2023)
Article
Computer Science, Software Engineering
Junqiu Zhu, Sizhe Zhao, Lu Wang, Yanning Xu, Ling-Qi Yan
Summary: This paper proposes an aggregated fur appearance model that accurately describes the optical behavior of a bunch of fur fibers using a single thick cylinder. By reducing the number of fur fibers and utilizing a lightweight neural network, the computational cost is significantly reduced while maintaining realistic fur appearance. The proposed method achieves almost identical results to the ground truth but performs 3.8x to 13.5x faster.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Software Engineering
Yuzhi Liang, Qi Song, Rui Wang, Yuchi Huo, Hujun Bao
Summary: This paper presents a new approach to combine simplifications of geometric and shading details in the level of detail (LOD) technique. The proposed method uses a discrete multiresolution representation that includes mesh and shader LODs. Two automatic algorithms are proposed to achieve the best simplifications of meshes and shaders at adaptively selected distances. The results demonstrate that the proposed mesh and shader LODs achieve better performance-quality tradeoffs compared to prior LOD representations that only consider simplified meshes or shaders.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2023)
Article
Computer Science, Software Engineering
Hui Zeng, Rong Zhao
Summary: A dual-mode virtual reality system is introduced based on the spatio-temporal perception characteristics of human vision. It can adaptively adjust the display spatial and temporal resolution according to the user's perceptual requirements for different display scenes, providing users with optimal visual perception quality.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2023)
Article
Robotics
Tong Nie, Jie Ma, Yuehua Zhao, Ziming Fan, Junjie Wen, Mengxuan Sun
Summary: In this study, a framework comprising geometry-guided instance-aware prior and multi-stage reconstruction networks is proposed to address the challenges in category-level object 6D pose estimation. The results of extensive experiments demonstrate a significant improvement in performance compared to previous methods.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Computer Science, Software Engineering
Rafal K. Mantiuk, Maliha Ashraf, Alexandre Chapiro
Summary: This paper proposes a unified contrast sensitivity function (CSF) that accounts for all major dimensions of the stimulus. By combining data from multiple studies, the function can accurately predict the data from these studies and explain the interactions between the dimensions of the CSF.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Artificial Intelligence
Jie Wang, Linhuang Yan, Qiaohe Yang, Minmin Yuan
Summary: A single-channel speech enhancement algorithm is proposed in this paper, utilizing guided spectrogram filtering based on the masking properties of the human auditory system to reduce background noise in the speech spectrogram, thereby improving speech clarity.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2021)
Article
Computer Science, Software Engineering
Berk Cebeci, Mehmet Bahadir Askin, Tolga K. Capin, Ufuk Celikcan
Summary: Despite advances in virtual reality technologies, extended VR sessions with head-mounted displays (HMDs) still face challenges in terms of comfort. In this study, a methodology using gaze-directed and visual saliency-guided paradigms for automatic stereo camera control in real-time interactive VR was proposed. The results showed that the gaze-directed approach outperformed the saliency-guided approach, both improving the overall depth feeling without hindering visual comfort in the tested virtual environments (VEs).
COMPUTERS & GRAPHICS-UK
(2024)
Article
Computer Science, Software Engineering
Luca Surace, Marek Wernikowski, Cara Tursun, Karol Myszkowski, Radoslaw Mantiuk, Piotr Didyk
Summary: This research aims to guide the training of foveated image reconstruction using generative adversarial networks (GANs) to reconstruct visually important image features. By conducting psychophysical experiments, building a dataset, and devising a training procedure, a strategy is proposed to penalize only perceptually important output deviations, resulting in improved image reconstruction quality.
ACM TRANSACTIONS ON APPLIED PERCEPTION
(2023)
Article
Computer Science, Software Engineering
Signe Sidwall Thygesen, Talha Bin Masood, Mathieu Linares, Vijay Natarajan, Ingrid Hotz
Summary: This pipeline presents an interactive visual analysis and exploration method for molecular electronic transition ensembles, providing support for theoretical chemists in comparing and characterizing electronic transitions. By combining automatic and interactive visual analysis, the pipeline offers multiple views to visualize and explore different aspects of the ensembles.
COMPUTER GRAPHICS FORUM
(2022)
Article
Computer Science, Software Engineering
Ailing Tan, Hongping Liao, Bozhi Zhang, Meijing Gao, Shiyu Li, Yang Bai, Zehao Liu
Summary: A new method is proposed in this paper to enhance infrared (IR) image details, effectively inhibiting image noise and improving image contrast while enhancing image details. The proposed method, called detail enhancement guided filter (DGIF), adds edge perception and detail regulation factors to the guided image filter (GIF) to overcome halo artifacts and noise amplification issues. Experimental results show that the DGIF outperforms other guided image filters and traditional IR image enhancement algorithms in terms of edge preservation, smoothing, and quantitative evaluation metrics. The algorithm effectively suppresses image noise while enhancing detail information, resulting in improved image contrast and better visual effect.
Article
Behavioral Sciences
Michele Fornaciai, Joonkoo Park
Summary: Previous studies suggest that numerosity processing involves at least two temporal stages, possibly reflecting initial feedforward processing and feedback signals from higher-order cortical areas. By utilizing different experimental methods, the results demonstrate that feedback processing is not strictly necessary for the perceptual segmentation that gives rise to perceived numerosity, but suggest that different stages of feedforward activity in early visual areas may be sufficient to create a numerosity representation.
Article
Robotics
Jiaqi Gu, Zhiyu Xiang, Yuwen Ye, Lingxuan Wang
Summary: A novel real-time pseudo-depth guided depth completion neural network is proposed to improve the output performance by guiding the network in three aspects using dense pseudo-depth maps. Experiments on the KITTI depth completion benchmark demonstrate that the model achieves state-of-the-art performance at the highest frame rate and shows significant improvements in downstream tasks.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Computer Science, Software Engineering
S. Frey
Summary: This paper introduces a new approach for generating grid layouts through optimization, allowing for level-of-detail presentation and exploration. By generating hierarchical grids and optimizing for homogeneity, the approach enables flexible exploration of large data collections.
COMPUTER GRAPHICS FORUM
(2022)