Search

Your search keyword '"Liu, Shixia"' showing total 550 results
Start Over Author "Liu, Shixia"
550 results on '"Liu, Shixia"'

51. A Unified Interactive Model Evaluation for Classification, Object Detection, and Instance Segmentation in Computer Vision

Author

Chen, Changjian, Guo, Yukai, Tian, Fengyuan, Liu, Shilong, Yang, Weikai, Wang, Zhaowei, Wu, Jing, Su, Hang, Pfister, Hanspeter, and Liu, Shixia

Abstract

Existing model evaluation tools mainly focus on evaluating classification models, leaving a gap in evaluating more complex models, such as object detection. In this paper, we develop an open-source visual analysis tool, Uni-Evaluator, to support a unified model evaluation for classification, object detection, and instance segmentation in computer vision. The key idea behind our method is to formulate both discrete and continuous predictions in different tasks as unified probability distributions. Based on these distributions, we develop 1) a matrix-based visualization to provide an overview of model performance; 2) a table visualization to identify the problematic data subsets where the model performs poorly; 3) a grid visualization to display the samples of interest. These visualizations work together to facilitate the model evaluation from a global overview to individual samples. Two case studies demonstrate the effectiveness of Uni-Evaluator in evaluating model performance and making informed improvements.

Published
2024
Full Text
View/download PDF

52. Cluster-Aware Grid Layout

Author

Zhou, Yuxing, Yang, Weikai, Chen, Jiashu, Chen, Changjian, Shen, Zhiyang, Luo, Xiaonan, Yu, Lingyun, and Liu, Shixia

Abstract

Grid visualizations are widely used in many applications to visually explain a set of data and their proximity relationships. However, existing layout methods face difficulties when dealing with the inherent cluster structures within the data. To address this issue, we propose a cluster-aware grid layout method that aims to better preserve cluster structures by simultaneously considering proximity, compactness, and convexity in the optimization process. Our method utilizes a hybrid optimization strategy that consists of two phases. The global phase aims to balance proximity and compactness within each cluster, while the local phase ensures the convexity of cluster shapes. We evaluate the proposed grid layout method through a series of quantitative experiments and two use cases, demonstrating its effectiveness in preserving cluster structures and facilitating analysis tasks.

Published
2024
Full Text
View/download PDF

53. Interactive Reweighting for Mitigating Label Quality Issues

Author

Yang, Weikai, Guo, Yukai, Wu, Jing, Wang, Zheng, Guo, Lan-Zhe, Li, Yu-Feng, and Liu, Shixia

Abstract

Label quality issues, such as noisy labels and imbalanced class distributions, have negative effects on model performance. Automatic reweighting methods identify problematic samples with label quality issues by recognizing their negative effects on validation samples and assigning lower weights to them. However, these methods fail to achieve satisfactory performance when the validation samples are of low quality. To tackle this, we develop Reweighter, a visual analysis tool for sample reweighting. The reweighting relationships between validation samples and training samples are modeled as a bipartite graph. Based on this graph, a validation sample improvement method is developed to improve the quality of validation samples. Since the automatic improvement may not always be perfect, a co-cluster-based bipartite graph visualization is developed to illustrate the reweighting relationships and support the interactive adjustments to validation samples and reweighting results. The adjustments are converted into the constraints of the validation sample improvement method to further improve validation samples. We demonstrate the effectiveness of Reweighter in improving reweighting results through quantitative evaluation and two case studies.

Published
2024
Full Text
View/download PDF

57. Diagnosing Ensemble Few-Shot Classifiers

Author

Yang, Weikai, primary, Ye, Xi, additional, Zhang, Xingxing, additional, Xiao, Lanxi, additional, Xia, Jiazhi, additional, Wang, Zhongyuan, additional, Zhu, Jun, additional, Pfister, Hanspeter, additional, and Liu, Shixia, additional

Published
2022
Full Text
View/download PDF

59. COBRA – A Visualization Solution to Monitor and Analyze Consumer Generated Medias

Author

Behal, Amit, Grace, Julia, Kato, Linda, Chen, Ying, Liu, Shixia, Cai, Weijia, Qian, Weihong, Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Nierstrasz, Oscar, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Sudan, Madhu, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Vardi, Moshe Y., Series editor, Weikum, Gerhard, Series editor, Salvendy, Gavriel, editor, and Smith, Michael J., editor

Published
2009
Full Text
View/download PDF

61. An LOD Model for Graph Visualization and Its Application in Web Navigation

Author

Liu, Shixia, Pan, Yue, Yang, Liping, Liu, Wenyin, Hutchison, David, editor, Kanade, Takeo, editor, Kittler, Josef, editor, Kleinberg, Jon M., editor, Mattern, Friedemann, editor, Mitchell, John C., editor, Naor, Moni, editor, Nierstrasz, Oscar, editor, Pandu Rangan, C., editor, Steffen, Bernhard, editor, Sudan, Madhu, editor, Terzopoulos, Demetri, editor, Tygar, Dough, editor, Vardi, Moshe Y., editor, Weikum, Gerhard, editor, Zhang, Yanchun, editor, Tanaka, Katsumi, editor, Yu, Jeffrey Xu, editor, Wang, Shan, editor, and Li, Minglu, editor

Published
2005
Full Text
View/download PDF

71. Preface

Author

Lee, Bongshin, primary, Miksch, Silvia, additional, Ynnerman, Anders, additional, Bezerianos, Anastasia, additional, Chen, Jian, additional, Chen, Wei, additional, Collins, Christopher, additional, Gleicher, Michael, additional, Groller, Eduard, additional, Lex, Alexander, additional, Preim, Bernhard, additional, Seo, Jinwook, additional, Westermann, Ruediger, additional, Yang, Jing, additional, Yuan, Xiaoru, additional, Shen, Han-Wei, additional, Fekete, Jean-Daniel, additional, and Liu, Shixia, additional

Published
2022
Full Text
View/download PDF

81. Visual Analysis of Neural Architecture Spaces for Summarizing Design Principles

Author

Yuan, Jun, Liu, Mengchen, Tian, Fengyuan, and Liu, Shixia

Abstract

Recent advances in artificial intelligence largely benefit from better neural network architectures. These architectures are a product of a costly process of trial-and-error. To ease this process, we develop ArchExplorer, a visual analysis method for understanding a neural architecture space and summarizing design principles. The key idea behind our method is to make the architecture space explainable by exploiting structural distances between architectures. We formulate the pairwise distance calculation as solving an all-pairs shortest path problem. To improve efficiency, we decompose this problem into a set of single-source shortest path problems. The time complexity is reduced from O(kn2N) to O(knN). Architectures are hierarchically clustered according to the distances between them. A circle-packing-based architecture visualization has been developed to convey both the global relationships between clusters and local neighborhoods of the architectures in each cluster. Two case studies and a post-analysis are presented to demonstrate the effectiveness of ArchExplorer in summarizing design principles and selecting better-performing architectures.

Published
2023
Full Text
View/download PDF

83. Visual Genealogy of Deep Neural Networks

Author

Wang, Qianwen, Yuan, Jun, Chen, Shuxin, Su, Hang, Qu, Huamin, Liu, Shixia, Wang, Qianwen, Yuan, Jun, Chen, Shuxin, Su, Hang, Qu, Huamin, and Liu, Shixia

Abstract

A comprehensive and comprehensible summary of existing deep neural networks (DNNs) helps practitioners understand the behaviour and evolution of DNNs, offers insights for architecture optimization, and sheds light on the working mechanisms of DNNs.However, this summary is hard to obtain because of the complexity and diversity of DNN architectures. To address this issue, we develop DNN Genealogy, an interactive visualization tool, to offer a visual summary of representative DNNs and their evolutionary relationships.DNN Genealogy enables users to learn DNNs from multiple aspects, including architecture, performance, and evolutionary relationships. Central to this tool is a systematic analysis and visualization of 66 representative DNNs based on our analysis of 140 papers. A directed acyclic graph is used to illustrate the evolutionary relationships among these DNNs and highlight the representative DNNs. A focus +context visualization is developed to orient users during their exploration. A set of network glyphs is used in the graph to facilitate the understanding and comparing of DNNs in the context of the evolution. Case studies demonstrate that DNN Genealogy provides helpful guidance in understanding, applying, and optimizing DNNs. DNN Genealogy is extensible and will continue to be updated to reflect future advances in DNNs.

Published
2020

84. Physiological and transcriptome analysis of γ-aminobutyric acid (GABA) in improving Gracilariopsis lemaneiformis stress tolerance at high temperatures

Author

Nianjun Xu, Liu Shixia, Zhang Jun, Xue Sun, and Chaoyang Hu

Subjects
chemistry.chemical_classification, Antioxidant, biology, medicine.medical_treatment, Metabolism, APX, Amino acid, Superoxide dismutase, chemistry, Biochemistry, Catalase, Heat shock protein, biology.protein, medicine, Proline, Agronomy and Crop Science
Abstract

Gracilariopsis lemaneiformis, an economically important red seaweed in China, is currently undergoing high-temperature (HT) stress due to global warming. To alleviate summertime HT stress, algae can acquire resistance through metabolism reprogramming and modulation of stress-related gene expression. γ-Aminobutyric acid (GABA), a non-protein four‑carbon amino acid, plays an important role in regulating various abiotic stresses. In the present study, GABA was applied to examine the effect on the tolerance of G. lemaneiformis against HT stress. It turned out that exogenous 100 mg/L GABA remarkably improved the algal relative growth rate (RGR) and phycobiliprotein (PBP) contents compared to those in the no-GABA treatment group under HT conditions. In addition, GABA significantly elevated the activities of antioxidant enzymes such as superoxide dismutase (SOD), peroxidase (POD) and catalase (CAT) to efficiently scavenge ROS, thereby reducing oxidative damage to the membranes of HT-stressed algae. Subsequently, a de novo transcriptomic assembly of G. lemaneiformis thalli was performed to analyze transcriptional responses to HT stress. GABA significantly upregulated the expression of genes encoding antioxidant enzymes (SOD, CAT, POD, PRXR and APX), triggered calmodulin (CaM) and calcium-dependent protein kinase (CDPK) signaling, and improved the expression of genes encoding heat shock proteins (HSP70s and HSP90s) during HT stress. Moreover, GABA enhanced the expression of genes involved in proline and polyamine (PA) synthesis, TCA cycle, amino acid metabolism and α-linolenic acid metabolism in G. lemaneiformis, which helped to maintain better energy supply and metabolic homeostasis under HT stress. Collectively, these data reveals the critical roles of GABA in antioxidant activation, molecular signaling and energy metabolism that coordinately contribute to the improvement of HT stress tolerance in G. lemaneiformis. These data will also be useful for breeding heat-tolerant G. lemaneiformis genotypes in the future.

Published
2021

87. Preserving Minority Structures in Graph Sampling

Author

Zhao, Ying, primary, Jiang, Haojin, additional, Chen, Qi'an, additional, Qin, Yaqi, additional, Xie, Huixuan, additional, Wu, Yitao, additional, Liu, Shixia, additional, Zhou, Zhiguang, additional, Xia, Jiazhi, additional, and Zhou, Fangfang, additional

Published
2021
Full Text
View/download PDF

90. Towards Better Caption Supervision for Object Detection.

Author

Chen, Changjian, Wu, Jing, Wang, Xiaohan, Xiang, Shouxing, Zhang, Song-Hai, Tang, Qifeng, and Liu, Shixia

Subjects
OBJECT recognition (Computer vision), SUPERVISION, DETECTORS
Abstract

As training high-performance object detectors requires expensive bounding box annotations, recent methods resort to free-available image captions. However, detectors trained on caption supervision perform poorly because captions are usually noisy and cannot provide precise location information. To tackle this issue, we present a visual analysis method, which tightly integrates caption supervision with object detection to mutually enhance each other. In particular, object labels are first extracted from captions, which are utilized to train the detectors. Then, the objects detected from images are fed into caption supervision for further improvement. To effectively loop users into the object detection process, a node-link-based set visualization supported by a multi-type relational co-clustering algorithm is developed to explain the relationships between the extracted labels and the images with detected objects. The co-clustering algorithm clusters labels and images simultaneously by utilizing both their representations and their relationships. Quantitative evaluations and a case study are conducted to demonstrate the efficiency and effectiveness of the developed method in improving the performance of object detectors. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

97. Exploring lekagul sensor events using rules, aggregations, and selections

Author

Cappers, Bram C.M., Schreck, Tobias, Fisher, Brian, Liu, Shixia, Algorithms, Geometry and Applications, and Security

Subjects
Multivariate statistics, Interaction, business.industry, Event (computing), Computer science, Multivariate events, 020207 software engineering, 02 engineering and technology, computer.software_genre, Regular expressions, Vehicle driving, Visualization, Data set, Data visualization, Event data, Sequence alignment, 0202 electrical engineering, electronic engineering, information engineering, Regular expression, Data mining, business, computer, Event visualization
Abstract

In this paper we demonstrate how we can study multivariate event sequences in the VAST Mini Challenge 1 data set using our system Eventpad, a notepad editor for event data. We illustrate the effectiveness of multivariate regular expressions, pattern aggregations, and selections to define custom events of interest, discover patterns within sequences, and study differences between sequences. Finally, we discuss our analysis process and summarize some patterns and anomalies we discovered in the data set.

Published
2018

98. Visual Genealogy of Deep Neural Networks

Author

Wang, Qianwen, Yuan, Jun, Chen, Shuxin, Su, Hang, Qu, Huamin, Liu, Shixia, Wang, Qianwen, Yuan, Jun, Chen, Shuxin, Su, Hang, Qu, Huamin, and Liu, Shixia

Abstract

A comprehensive and comprehensible summary of existing deep neural networks (DNNs) helps practitioners understand the behaviour and evolution of DNNs, offers insights for architecture optimization, and sheds light on the working mechanisms of DNNs.However, this summary is hard to obtain because of the complexity and diversity of DNN architectures. To address this issue, we develop DNN Genealogy, an interactive visualization tool, to offer a visual summary of representative DNNs and their evolutionary relationships.DNN Genealogy enables users to learn DNNs from multiple aspects, including architecture, performance, and evolutionary relationships. Central to this tool is a systematic analysis and visualization of 66 representative DNNs based on our analysis of 140 papers. A directed acyclic graph is used to illustrate the evolutionary relationships among these DNNs and highlight the representative DNNs. A focus +context visualization is developed to orient users during their exploration. A set of network glyphs is used in the graph to facilitate the understanding and comparing of DNNs in the context of the evolution. Case studies demonstrate that DNN Genealogy provides helpful guidance in understanding, applying, and optimizing DNNs. DNN Genealogy is extensible and will continue to be updated to reflect future advances in DNNs.

Published
2019

Catalog

Books, media, physical & digital resources
See catalog results