ORCID Profile
0000-0002-9640-6472
Current Organisation
University of Sydney
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
In Research Link Australia (RLA), "Research Topics" refer to ANZSRC FOR and SEO codes. These topics are either sourced from ANZSRC FOR and SEO codes listed in researchers' related grants or generated by a large language model (LLM) based on their publications.
Artificial Intelligence and Image Processing | Pattern Recognition and Data Mining | Computer Vision | Machine learning | Semi- and unsupervised learning | Field robotics | Statistics | Intelligent robotics | Computer vision | Knowledge Representation and Machine Learning | Applied Statistics | Natural Resource Management | Adversarial machine learning | Artificial intelligence
Expanding Knowledge in the Information and Computing Sciences | Film and Video Services (excl. Animation and Computer Generated Imagery) | Information Processing Services (incl. Data Entry and Capture) | Environmental Management Systems | Expanding Knowledge in the Medical and Health Sciences | Precious (Noble) Metal Ore Exploration | Mining Land and Water Management |
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 02-2017
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2017
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2021
Publisher: arXiv
Date: 2022
Publisher: ACM
Date: 10-08-2015
Publisher: Springer International Publishing
Date: 2019
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 02-2018
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 08-2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 09-2016
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 06-2018
Publisher: arXiv
Date: 2022
Publisher: IEEE
Date: 04-2021
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 03-2018
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2019
Publisher: ACM
Date: 14-08-2022
Publisher: Association for Computing Machinery (ACM)
Date: 17-02-2016
DOI: 10.1145/2819000
Abstract: Editing faces in videos is a popular yet challenging task in computer vision and graphics that encompasses various applications, including facial attractiveness enhancement, makeup transfer, face replacement, and expression manipulation. Directly applying the existing warping methods to video face editing has the major problem of temporal incoherence in the synthesized videos, which cannot be addressed by simply employing face tracking techniques or manual interventions, as it is difficult to eliminate the subtly temporal incoherence of the facial feature point localizations in a video sequence. In this article, we propose a temporal-spatial-smooth warping (TSSW) method to achieve a high temporal coherence for video face editing. TSSW is based on two observations: (1) the control lattices are critical for generating warping surfaces and achieving the temporal coherence between consecutive video frames, and (2) the temporal coherence and spatial smoothness of the control lattices can be simultaneously and effectively preserved. Based upon these observations, we impose the temporal coherence constraint on the control lattices on two consecutive frames, as well as the spatial smoothness constraint on the control lattice on the current frame. TSSW calculates the control lattice (in either the horizontal or vertical direction) by updating the control lattice (in the corresponding direction) on its preceding frame, i.e., minimizing a novel energy function that unifies a data-driven term, a smoothness term, and feature point constraints. The contributions of this article are twofold: (1) we develop TSSW, which is robust to the subtly temporal incoherence of the facial feature point localizations and is effective to preserve the temporal coherence and spatial smoothness of the control lattices for editing faces in videos, and (2) we present a new unified video face editing framework that is capable for improving the performances of facial attractiveness enhancement, makeup transfer, face replacement, and expression manipulation.
Publisher: American Physical Society (APS)
Date: 27-05-2021
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2023
Publisher: IEEE
Date: 07-2017
DOI: 10.1109/CVPR.2017.15
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2020
Publisher: American Physical Society (APS)
Date: 22-07-2018
Publisher: IEEE
Date: 04-2014
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 10-2022
Publisher: International Joint Conferences on Artificial Intelligence Organization
Date: 08-2019
Abstract: Positive and Unlabeled (PU) learning aims to learn a binary classifier from only positive and unlabeled training data. The state-of-the-art methods usually formulate PU learning as a cost-sensitive learning problem, in which every unlabeled ex le is simultaneously treated as positive and negative with different class weights. However, the ground-truth label of an unlabeled ex le should be unique, so the existing models inadvertently introduce the label noise which may lead to the biased classifier and deteriorated performance. To solve this problem, this paper proposes a novel algorithm dubbed as "Positive and Unlabeled learning with Label Disambiguation'' (PULD). We first regard all the unlabeled ex les in PU learning as ambiguously labeled as positive and negative, and then employ the margin-based label disambiguation strategy, which enlarges the margin of classifier response between the most likely label and the less likely one, to find the unique ground-truth label of each unlabeled ex le. Theoretically, we derive the generalization error bound of the proposed method by analyzing its Rademacher complexity. Experimentally, we conduct intensive experiments on both benchmark and real-world datasets, and the results clearly demonstrate the superiority of the proposed PULD to the existing PU learning approaches.
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2023
Publisher: ACM
Date: 26-10-2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2021
Publisher: Springer Science and Business Media LLC
Date: 09-01-2021
Publisher: American Physical Society (APS)
Date: 24-06-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2021
Publisher: IEEE
Date: 07-2019
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 04-2019
Publisher: IEEE
Date: 06-2022
Publisher: IEEE
Date: 05-2022
Publisher: IEEE
Date: 07-2014
Publisher: American Physical Society (APS)
Date: 14-03-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 09-2020
Publisher: Elsevier BV
Date: 10-2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 03-2016
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 03-2021
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 05-2018
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2016
Publisher: IEEE
Date: 10-2021
Publisher: Springer International Publishing
Date: 2018
Publisher: IEEE
Date: 10-2021
Publisher: ACM
Date: 14-08-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 08-2015
Publisher: International Joint Conferences on Artificial Intelligence Organization
Date: 07-2018
Abstract: It is NP-complete to find non-negative factors W and H with fixed rank r from a non-negative matrix X by minimizing ||X-WH^Τ ||^2. Although the separability assumption (all data points are in the conical hull of the extreme rows) enables polynomial-time algorithms, the computational cost is not affordable for big data. This paper investigates how the power of quantum computation can be capitalized to solve the non-negative matrix factorization with the separability assumption (SNMF) by devising a quantum algorithm based on the ide-and-conquer anchoring (DCA) scheme [Zhou et al., 2013]. The design of quantum DCA (QDCA) is challenging. In the ide step, the random projections in DCA is completed by a quantum algorithm for linear operations, which achieves the exponential speedup. We then devise a heuristic post-selection procedure which extracts the information of anchors stored in the quantum states efficiently. Under a plausible assumption, QDCA performs efficiently, achieves the quantum speedup, and is beneficial for high dimensional problems.
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 10-2015
Publisher: IEEE
Date: 06-2018
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 06-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 03-2018
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 04-2021
Publisher: IEEE
Date: 06-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 09-2018
Publisher: American Physical Society (APS)
Date: 17-11-2021
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 09-2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 11-2019
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 06-2020
Publisher: ACM
Date: 17-10-2022
Publisher: IEEE
Date: 06-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 12-2020
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 04-2022
Publisher: IEEE
Date: 26-09-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 12-2022
Publisher: ACM
Date: 07-11-2022
Publisher: International Joint Conferences on Artificial Intelligence Organization
Date: 08-2017
Abstract: Transfer learning transfers knowledge across domains to improve the learning performance. Since feature structures generally represent the common knowledge across different domains, they can be transferred successfully even though the labeling functions across domains differ arbitrarily. However, theoretical justification for this success has remained elusive. In this paper, motivated by self-taught learning, we regard a set of bases as a feature structure of a domain if the bases can (approximately) reconstruct any observation in this domain. We propose a general analysis scheme to theoretically justify that if the source and target domains share similar feature structures, the source domain feature structure is transferable to the target domain, regardless of the change of the labeling functions across domains. The transferred structure is interpreted to function as a regularization matrix which benefits the learning process of the target domain task. We prove that such transfer enables the corresponding learning algorithms to be uniformly stable. Specifically, we illustrate the existence of feature structure transfer in two well-known transfer learning settings: domain adaptation and learning to learn.
Publisher: Journal of Graph Algorithms and Applications
Date: 2022
DOI: 10.7155/JGAA.00603
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 05-2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2022
Publisher: Springer International Publishing
Date: 2018
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 05-2017
Publisher: MDPI AG
Date: 31-03-2022
DOI: 10.3390/FRACTALFRACT6040197
Abstract: The large proportion of asymptomatic patients is the major cause leading to the COVID-19 pandemic which is still a significant threat to the whole world. A six-dimensional ODE system (SEIAQR epidemical model) is established to study the dynamics of COVID-19 spreading considering infection by exposed, infected, and asymptomatic cases. The basic reproduction number derived from the model is more comprehensive including the contribution from the exposed, infected, and asymptomatic patients. For this more complex six-dimensional ODE system, we investigate the global and local stability of disease-free equilibrium, as well as the endemic equilibrium, whereas most studies overlooked asymptomatic infection or some other virus transmission features. In the sensitivity analysis, the parameters related to the asymptomatic play a significant role not only in the basic reproduction number R0. It is also found that the asymptomatic infection greatly affected the endemic equilibrium. Either in completely eradicating the disease or achieving a more realistic goal to reduce the COVID-19 cases in an endemic equilibrium, the importance of controlling the asymptomatic infection should be emphasized. The three-dimensional phase diagrams demonstrate the convergence point of the COVID-19 spreading under different initial conditions. In particular, massive infections will occur as shown in the phase diagram quantitatively in the case R0 . Moreover, two four-dimensional contour maps of Rt are given varying with different parameters, which can offer better intuitive instructions on the control of the pandemic by adjusting policy-related parameters.
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 08-2016
Publisher: IEEE
Date: 06-2021
Publisher: American Astronomical Society
Date: 29-06-2017
Publisher: American Physical Society (APS)
Date: 27-08-2021
Publisher: Institution of Engineering and Technology (IET)
Date: 31-01-2019
Publisher: IEEE
Date: 11-07-2021
Publisher: IEEE
Date: 06-2021
Publisher: Springer International Publishing
Date: 2021
Publisher: IEEE
Date: 10-2021
Publisher: Springer International Publishing
Date: 2018
Publisher: American Physical Society (APS)
Date: 11-07-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 08-2016
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 12-2022
Publisher: IEEE
Date: 06-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 06-2020
Publisher: International Joint Conferences on Artificial Intelligence Organization
Date: 07-2018
Abstract: Hashing is becoming increasingly popular for approximate nearest neighbor searching in massive databases due to its storage and search efficiency. Recent supervised hashing methods, which usually construct semantic similarity matrices to guide hash code learning using label information, have shown promising results. However, it is relatively difficult to capture and utilize the semantic relationships between points in unsupervised settings. To address this problem, we propose a novel unsupervised deep framework called Semantic Structure-based unsupervised Deep Hashing (SSDH). We first empirically study the deep feature statistics, and find that the distribution of the cosine distance for point pairs can be estimated by two half Gaussian distributions. Based on this observation, we construct the semantic structure by considering points with distances obviously smaller than the others as semantically similar and points with distances obviously larger than the others as semantically dissimilar. We then design a deep architecture and a pair-wise loss function to preserve this semantic structure in Hamming space. Extensive experiments show that SSDH significantly outperforms current state-of-the-art methods.
Publisher: IEEE
Date: 06-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 10-2015
Publisher: Elsevier BV
Date: 04-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 03-2016
Publisher: International Joint Conferences on Artificial Intelligence Organization
Date: 08-2017
Abstract: Transfer learning aims to improve the performance of target learning task by leveraging information (or transferring knowledge) from other related tasks. Recently, transfer distance metric learning (TDML) has attracted lots of interests, but most of these methods assume that feature representations for the source and target learning tasks are the same. Hence, they are not suitable for the applications, in which the data are from heterogeneous domains (feature spaces, modalities and even semantics). Although some existing heterogeneous transfer learning (HTL) approaches is able to handle such domains, they lack flexibility in real-world applications, and the learned transformations are often restricted to be linear. We therefore develop a general and flexible heterogeneous TDML (HTDML) framework based on the knowledge fragment transfer strategy. In the proposed HTDML, any (linear or nonlinear) distance metric learning algorithms can be employed to learn the source metric beforehand. Then a set of knowledge fragments are extracted from the pre-learned source metric to help target metric learning. In addition, either linear or nonlinear distance metric can be learned for the target domain. Extensive experiments on both scene classification and object recognition demonstrate superiority of the proposed method.
Publisher: MIT Press - Journals
Date: 10-2016
DOI: 10.1162/NECO_A_00872
Abstract: The k-dimensional coding schemes refer to a collection of methods that attempt to represent data using a set of representative k-dimensional vectors and include nonnegative matrix factorization, dictionary learning, sparse coding, k-means clustering, and vector quantization as special cases. Previous generalization bounds for the reconstruction error of the k-dimensional coding schemes are mainly dimensionality-independent. A major advantage of these bounds is that they can be used to analyze the generalization error when data are mapped into an infinite- or high-dimensional feature space. However, many applications use finite-dimensional data features. Can we obtain dimensionality-dependent generalization bounds for k-dimensional coding schemes that are tighter than dimensionality-independent bounds when data are in a finite-dimensional feature space? Yes. In this letter, we address this problem and derive a dimensionality-dependent generalization bound for k-dimensional coding schemes by bounding the covering number of the loss function class induced by the reconstruction error. The bound is of order [Formula: see text], where m is the dimension of features, k is the number of the columns in the linear implementation of coding schemes, and n is the size of s le, [Formula: see text] when n is finite and [Formula: see text] when n is infinite. We show that our bound can be tighter than previous results because it avoids inducing the worst-case upper bound on k of the loss function. The proposed generalization bound is also applied to some specific coding schemes to demonstrate that the dimensionality-dependent bound is an indispensable complement to the dimensionality-independent generalization bounds.
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 11-2020
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 08-2017
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 11-04-2021
DOI: 10.36227/TECHRXIV.14380697
Abstract: A major gap between few-shot and many-shot learning is the data distribution empirically observed by the model during training. In few-shot learning, the learned model can easily become over-fitted based on the biased distribution formed by only a few training ex les, while the ground-truth data distribution is more accurately uncovered in many-shot learning to learn a well-generalized model. In this paper, we propose to calibrate the distribution of these few-s le classes to be more unbiased to alleviate such an over-fitting problem. The distribution calibration is achieved by transferring statistics from the classes with sufficient ex les to those few-s le classes. After calibration, an adequate number of ex les can be s led from the calibrated distribution to expand the inputs to the classifier. Extensive experiments on three datasets, miniImageNet, tieredImageNet, and CUB, show that a simple linear classifier trained using the features s led from our calibrated distribution can outperform the state-of-the-art accuracy by a large margin. We also establish a generalization error bound for the proposed distribution-calibration-based few-shot learning, which consists of the distribution assumption error, the distribution approximation error, and the estimation error. This generalization error bound theoretically justifies the effectiveness of the proposed method.
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2023
Publisher: ACM
Date: 12-10-2020
Publisher: International Joint Conferences on Artificial Intelligence Organization
Date: 07-2018
Abstract: Distance metric learning (DML) has been demonstrated to be successful and essential in erse applications. Transfer metric learning (TML) can help DML in the target domain with limited label information by utilizing information from some related source domains. The heterogeneous TML (HTML), where the feature representations vary from the source to the target domain, is general and challenging. However, current HTML approaches are usually conducted in a batch manner and cannot handle sequential data. This motivates the proposed online HTML (OHTML) method. In particular, the distance metric in the source domain is pre-trained using some existing DML algorithms. To enable knowledge transfer, we assume there are large amounts of unlabeled corresponding data that have representations in both the source and target domains. By enforcing the distances (between these unlabeled s les) in the target domain to agree with those in the source domain under the manifold regularization theme, we learn an improved target metric. We formulate the problem in the online setting so that the optimization is efficient and the model can be adapted to new coming data. Experiments in erse applications demonstrate both effectiveness and efficiency of the proposed method.
Publisher: Elsevier BV
Date: 06-2022
Publisher: American Physical Society (APS)
Date: 05-08-2019
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2022
Publisher: IOP Publishing
Date: 02-2021
Abstract: The hybrid quantum–classical learning scheme provides a prominent way to achieve quantum advantages on near-term quantum devices. A concrete ex le toward this goal is the quantum neural network (QNN), which has been developed to accomplish various supervised learning tasks such as classification and regression. However, there are two central issues that remain obscure when QNN is exploited to accomplish classification tasks. First, a quantum classifier that can well balance the computational cost such as the number of measurements and the learning performance is unexplored. Second, it is unclear whether quantum classifiers can be applied to solve certain problems that outperform their classical counterparts. Here we devise a Grover-search based quantum learning scheme (GBLS) to address the above two issues. Notably, most existing QNN-based quantum classifiers can be seamlessly embedded into the proposed scheme. The key insight behind our proposal is reformulating the classification tasks as the search problem. Numerical simulations exhibit that GBLS can achieve comparable performance with other quantum classifiers under various noise settings, while the required number of measurements is dramatically reduced. We further demonstrate a potential quantum advantage of GBLS over classical classifiers in the measure of query complexity. Our work provides guidance to develop advanced quantum classifiers on near-term quantum devices and opens up an avenue to explore potential quantum advantages in various classification tasks.
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 08-2019
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 08-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 09-2014
Publisher: IEEE
Date: 06-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 04-2020
Publisher: Springer International Publishing
Date: 03-04-2019
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 06-2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 02-2017
Publisher: Elsevier BV
Date: 12-2023
Publisher: IEEE
Date: 06-2019
Publisher: Elsevier BV
Date: 07-2019
Publisher: Springer International Publishing
Date: 2020
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2019
Publisher: IEEE
Date: 06-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 06-2019
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 11-2019
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 10-2018
Start Date: 07-2023
End Date: 06-2026
Amount: $419,886.00
Funder: Australian Research Council
View Funded ActivityStart Date: 07-2023
End Date: 06-2026
Amount: $478,994.00
Funder: Australian Research Council
View Funded ActivityStart Date: 06-2019
End Date: 05-2022
Amount: $387,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 06-2018
End Date: 06-2021
Amount: $392,893.00
Funder: Australian Research Council
View Funded ActivityStart Date: 08-2022
End Date: 03-2026
Amount: $405,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 06-2023
End Date: 06-2027
Amount: $764,534.00
Funder: Australian Research Council
View Funded ActivityStart Date: 08-2020
End Date: 08-2025
Amount: $3,973,202.00
Funder: Australian Research Council
View Funded Activity