Tsung-Han Wu

Tsung-Han Wu

Auto-generated profile, based on publicly available information

Tsung-Han Wu

National Taiwan University

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 March 2018 in IEEE Transactions on Network Science and Engineering
Verified
Journal Article
Closed access
TWTsunghan Wu
Tsunghan Wu
CCCheng-Shang Chang
Cheng-Shang Chang
WLWanjiun Liao
Wanjiun Liao
Structural network analysis, including node ranking, community detection, and link prediction, has received a lot of attention lately. In the literature, most works focused on the structural analysis of a single network. In this paper, we are particularly interested in how the network structure evolves over time. For this, we propose a general framework to track, model, and predict the dynamic network structures. Unlike some recent works that directly tracks the adjacency matrices of the networks, our framework utilizes the spectral graph theory to track the latent feature vectors obtained by a low-rank eigendecomposition of the Laplacian matrices of the networks. We then use the Finite Impulse Response (FIR) filter to model the evolution of the latent feature vector of each node. By solving a ridge regression problem, the parameters of the FIR filter can be learned and used for predicting the future network structures, including node ranking, community detection, and link prediction. To test the effectiveness of our framework, we perform various experiments based on our synthetic datasets and three real-world datasets. Our experimental results show that our framework is very effective in tracking latent feature vectors and predicting future network structures.

1 September 2011 in Expert Systems with Applications
Verified
Journal Article
Closed access
CLChunshien Li
Chunshien Li
TWTsunghan Wu
Tsunghan Wu

2 October 2023
Verified
Conference Paper
Open access
TWTsung-Han Wu
Tsung-Han Wu
HSHung-Ting Su
Hung-Ting Su
SCShang-Tse Chen
Shang-Tse Chen
WHWinston H. Hsu
Winston H. Hsu
We introduce a new learning framework, Fair Robust Active Learning (FRAL), generalizing conventional active learning to fair and adversarial robust scenarios. This framework enables us to achieve fair-performance and fairrobustness with limited labeled data, which is essential for various annotation-expensive visual applications with safety-critical needs. However, existing fairness-aware data selection strategies face two challenges when applied to the FRAL framework: they are either ineffective under severe data imbalance or inefficient due to huge computations of adversarial training. To address these issues, we develop a novel Joint INconsistency (JIN) method that exploits prediction inconsistencies between benign and adversarial inputs and between standard and robust models. By leveraging these two types of easy-to-compute inconsistencies simultaneously, JIN can identify valuable samples that contribute more to fairness gains and class imbalance mitigation in both standard and adversarial robust settings. Extensive experiments on diverse datasets and sensitive groups demonstrate that our approach outperforms existing active data selection baselines, achieving fair-performance and fairrobustness under white-box PGD attacks.

24 April 2020 in arXiv
Verified
Preprint
Open access
PDF file
YHYu-Kai Huang
Yu-Kai Huang
YLYueh-Cheng Liu
Yueh-Cheng Liu
TWTsung-Han Wu
Tsung-Han Wu
HSHung-Ting Su
Hung-Ting Su
WHWinston H. Hsu
Winston H. Hsu
The performance of image based stereo estimation suffers from lighting variations, repetitive patterns and homogeneous appearance. Moreover, to achieve good performance, stereo supervision requires sufficient densely-labeled data, which are hard to obtain. In this work, we leverage small amount of data with very sparse but accurate disparity cues from LiDAR to bridge the gap. We propose a novel sparsity expansion technique to expand the sparse cues concerning RGB images for local feature enhancement. The feature enhancement method can be easily applied to any stereo estimation algorithms with cost volume at the test stage. Extensive experiments on stereo datasets demonstrate the effectiveness and robustness across different backbones on domain adaption and self-supervision scenario. Our sparsity expansion method outperforms previous methods in terms of disparity by more than 2 pixel error on KITTI Stereo 2012 and 3 pixel error on KITTI Stereo 2015. Our approach significantly boosts the existing state-of-the-art stereo algorithms with extremely sparse cues.

22 August 2021 in arXiv
Verified
Preprint
Open access
PDF file
YHYu-Kai Huang
Yu-Kai Huang
YLYueh-Cheng Liu
Yueh-Cheng Liu
TWTsung-Han Wu
Tsung-Han Wu
HSHung-Ting Su
Hung-Ting Su
YCYu-Cheng Chang
Yu-Cheng Chang
WHWinston H. Hsu
Winston H. Hsu
Dense depth estimation plays a key role in multiple applications such as robotics, 3D reconstruction, and augmented reality. While sparse signal, e.g., LiDAR and Radar, has been leveraged as guidance for enhancing dense depth estimation, the improvement is limited due to its low density and imbalanced distribution. To maximize the utility from the sparse source, we propose $S^3$ technique, which expands the depth value from sparse cues while estimating the confidence of expanded region. The proposed $S^3$ can be applied to various guided depth estimation approaches and trained end-to-end at different stages, including input, cost volume and output. Extensive experiments demonstrate the effectiveness, robustness, and flexibility of the $S^3$ technique on LiDAR and Radar signal.

23 November 2023
Verified
Conference Paper
Closed access
TWTsung-Han Wu
Tsung-Han Wu
SWShun-Min Wang
Shun-Min Wang
Wen-Wei ChangWen-Wei Chang
Wen-Wei Chang
In the face of high-risk and unpredictable ocean conditions, it is impractical to send divers for underwater exploration. This research aims to develop a small and highly nimble underwater manipulator system. Additionally, an Underwater Vision Sensing System was used to distinguish the trajectory of the target through cameras, and to perform target grabs and retrieval work using unmanned underwater manipulator. The underwater multi-axis manipulator in this study used spatial coordinate transformation to calculate the forward and inverse kinematics of the manipulator. In addition, the appropriate movement posture that matches the manipulator was selected through condition filters, enabling the manipulators to achieve desired target position. In terms of control, LabView, a graphical programming platform, was used to design the user interface, and the gyroscope measured the angle error of the manipulators and fed it back to the controllers. There are two types of servo motor controllers tested in this research: a feedforward proportional-integral-derivative (PID) controller and a Gain-Scheduled PID controller. The gyroscopic measurement results showed that using Gain-Scheduled PID Controller provides higher accuracy than the others controller. This study applies existing underwater manipulator control theories to develop and modify the motion-controlling system through simulation and actual operation. The contribution of this research will expand the application of underwater exploration in both academic and industrial fields.