Article
Engineering, Electrical & Electronic
Xinyu Lin, Yingjie Zhou, Xun Zhang, Yipeng Liu, Ce Zhu
Summary: Binary feature descriptors are widely used in visual measurement tasks, but they may not perform well for long-term visual measurement tasks due to their sensitivity to illumination variations. This study presents an illumination-insensitive binary (IIB) descriptor that leverages the local inter-patch invariance exhibited in multiple spatial granularities to deal with unfavorable illumination variations. Numerical experiments demonstrate that the proposed IIB descriptor outperforms state-of-the-art binary descriptors and some float descriptors.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Chemistry, Physical
Claudio Zeni, Kevin Rossi, Aldo Glielmo, Stefano de Gironcoli
Summary: The study shows that a simple linear regression framework performs comparably with advanced machine learning methods in predicting formation energies and atomic forces, while also achieving improved computational efficiency. By utilizing principal component analysis and least absolute shrinkage operator regression, it is possible to construct descriptors that are significantly smaller yet maintain similar or improved accuracy, with shared features across multiple datasets.
JOURNAL OF CHEMICAL PHYSICS
(2021)
Review
Chemistry, Physical
E. Santos
Summary: The article discusses the limitations and risks of using thermodynamic and electronic properties as descriptors to obtain simple correlations in electro-catalysis.
CURRENT OPINION IN ELECTROCHEMISTRY
(2023)
Article
Chemistry, Analytical
Rasha Shoitan, Mona M. Moussa, Sawsan Morkos Gharghory, Heba A. Elnemr, Young-Im Cho, Mohamed S. Abdallah
Summary: Surveillance cameras have become widespread over the past decade and their usage is expected to increase further. Browsing and analyzing these surveillance videos effectively is crucial in surveillance applications. A video synopsis approach has been proposed recently to reduce video duration by rearranging objects. However, summarizing crowded videos is challenging. Different clustering and query methods have been introduced to generate video synopsis based on attributes like color, size, class, and motion. This work presents a user-defined query synopsis video using motion and specific appearance features, assisting in retrieving people who meet certain criteria and generating a short video. Evaluation shows high precision, recall, and F1 score for the retrieval process, as well as significant reduction in video length and preservation of interactions in the synopsis video.
Article
Astronomy & Astrophysics
A. L. Patel, J. S. Urquhart, A. Y. Yang, T. J. T. Moore, K. M. Menten, M. A. Thompson, M. G. Hoare, T. Irabor, S. L. Breen, M. D. Smith
Summary: This study uses archival high-frequency continuum data to expand the search for Hypercompact H ii regions and determine the conditions at which they appear. The researchers used 23 GHz continuum data towards methanol masers and identified 49 H ii regions, 47 of which are embedded in dense clumps. They also identified 13 methanol maser sites that are coincident with radio sources with a steep positive spectral index, suggesting they are good HC H ii region candidates.
MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY
(2023)
Article
Computer Science, Artificial Intelligence
Run Li, Yucheng Shi, Yahong Han, Yunfeng Shao, Meiyu Qi, Bingshuai Li
Summary: Entropy search and its derivative methods are Bayesian Optimization methods that actively explore black-box functions by maximizing information gain. However, existing entropy search methods are limited in high dimensional problems and lack a mechanism to exclude redundant dimensions. In this work, Active Compact Entropy Search (AcCES) is proposed to address these issues by considering the correlation between dimensions and compressing redundant dimensions. Experimental results show that AcCES achieves higher query efficiency and optimal results in convergence compared to existing entropy search methods.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Psychology, Experimental
Anna Nowakowska, Alasdair D. F. Clarke, Josephine Reuther, Amelia R. Hunt
Summary: This study compares eye movement strategies across different stimulus sets and finds that eye movements are guided by expected information gain. When searching for a simple target based on orientation, people vary in their eye movements and often look at locations where peripheral vision is sufficient for determining the target's presence. In contrast, when searching for a target based on identity, eye movements are similar and directed towards locations where central vision is most needed.
JOURNAL OF EXPERIMENTAL PSYCHOLOGY-GENERAL
(2023)
Article
Computer Science, Artificial Intelligence
Andrea Migliorati, Attilio Fiandrotti, Gianluca Francini, Riccardo Leonardi
Summary: LDVS is a learnable binary local descriptor designed for matching natural images within the MPEG CDVS framework. Through experiments, LDVS descriptors have shown favorable performance in image patch matching and have a moderate parameters count for operations on mobile devices.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Physics, Applied
F. Mitri, A. De Iacovo, S. De Santis, C. Giansante, D. Spirito, G. Sotgiu, L. Colace
Summary: The study introduces a compact, low-cost sensor for explosive detection based on solid-state PbS quantum dot solids cast on a silicon substrate, with the capability to detect nitrobenzene vapor at room temperature. The system can be implemented with off-the-shelf electronics and does not require additional laboratory equipment for operation, making it suitable for deployment in distributed sensor networks.
APPLIED PHYSICS LETTERS
(2021)
Article
Remote Sensing
Haodong Xiang, Wenzhong Shi, Wenzheng Fan, Pengxin Chen, Sheng Bao, Mingyan Nie
Summary: In this paper, a fast and compact loop closure detection method is proposed based on comprehensive descriptors and machine learning for indoor LiDAR mobile mapping, achieving reliable and precise results. By feeding specific descriptor values into a machine learning model and using a loop candidate verification strategy, the proposed method shows superior performance in precision and recall rate.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Astronomy & Astrophysics
A. Ferre-Mateu, M. Mezcua, R. S. Barrows
Summary: A study on the active BH activity in low-mass compact galaxies reveals that some compact elliptical galaxies host AGN, and these galaxies tend to be overmassive in the BH-galaxy scaling relations, supporting a stripping origin for the majority of these objects.
MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY
(2021)
Article
Computer Science, Artificial Intelligence
Baofu Fang, Gaofei Mei, Xiaohui Yuan, Le Wang, Zaijun Wang, Junyang Wang
Summary: This paper introduces a novel SLAM technology using RGB and depth images combined with knowledge graph and semantic descriptor to effectively handle dynamic environments, significantly improving the accuracy of robot tracking and positioning in medical facilities. The seamless integration of knowledge graph and semantic descriptor helps eliminate dynamic objects, enhancing operational efficiency in hospitals.
PATTERN RECOGNITION
(2021)
Article
Engineering, Electrical & Electronic
Bolin Chen, Zhao Wang, Binzhe Li, Shiqi Wang, Yan Ye
Summary: In this paper, a method for compactly representing the nonlinear dynamics of talking face videos is proposed. The method is suitable for ultra-low bandwidth video communication and demonstrates superior performance in video quality reconstruction.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Astronomy & Astrophysics
Leo Tsukada, Prathamesh Joshi, Shomik Adhicary, Richard George, Andre Guimaraes, Chad Hanna, Ryan Magee, Aaron Zimmerman, Pratyusava Baral, Amanda Baylor, Kipp Cannon, Sarah Caudill, Bryce Cousins, Jolien D. E. Creighton, Becca Ewing, Heather Fong, Patrick Godwin, Reiko Harada, Yun-Jing Huang, Rachael Huxford, James Kennington, Soichiro Kuwahara, Alvin K. Y. Li, Duncan Meacher, Cody Messick, Soichiro Morisaki, Debnandini Mukherjee, Wanting Niu, Alex Pace, Cort Posnansky, Anarya Ray, Surabhi Sachdev, Shio Sakon, Divya Singh, Ron Tapia, Takuya Tsutsui, Koh Ueno, Aaron Viets, Leslie Wade, Madeline Wade
Summary: This study discusses new features developed for the GstLAL-based inspiral pipeline, which have been shown to improve the pipeline's sensitivity by approximately 15% to 20% through an injection study, paving the way for further multimessenger observations during the upcoming observing run.
Article
Multidisciplinary Sciences
Jacob A. Westerberg, Elizabeth A. Sigworth, Jeffrey D. Schall, Alexander Maier
Summary: The research indicates that attention plays a significant role in feature selectivity during visual search, especially in enhancing feature tuning in deep neural layers, which directly affects attentional functions and behavioral outcomes.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)