- School of Mathematics and Statistics - Research Publications
School of Mathematics and Statistics - Research Publications
Permanent URI for this collection
19 results
Filters
Reset filtersSettings
Statistics
Citations
Search Results
Now showing
1 - 10 of 19
-
ItemNo Preview AvailableCombating Noisy Labels with Sample Selection by Mining High-Discrepancy ExamplesXia, X ; Han, B ; Zhan, Y ; Yu, J ; Gong, M ; Gong, C ; Liu, T (IEEE, 2023-01-01)
-
ItemNo Preview AvailableGenerating Dynamic Kernels via Transformers for Lane DetectionChen, Z ; Liu, Y ; Gong, M ; Du, B ; Qian, G ; Smith-Miles, K (IEEE, 2023-01-01)
-
ItemNo Preview AvailableKnowledge Distillation for Feature Extraction in Underwater VSLAMYang, J ; Gong, M ; Nair, G ; Lee, JH ; Monty, J ; Pu, Y (IEEE, 2023-01-01)
-
ItemNo Preview AvailableUnpaired Image-to-Image Translation with Shortest Path RegularizationXie, S ; Xu, Y ; Gong, M ; Zhang, K (IEEE, 2023-06)
-
ItemNo Preview AvailableAdaptive Local-Component-aware Graph Convolutional Network for One-shot Skeleton-based Action RecognitionZhu, A ; Ke, Q ; Gong, M ; Bailey, J (IEEE COMPUTER SOC, 2023)
-
ItemNo Preview AvailableProgressive Video Summarization via Multimodal Self-supervised LearningLi, H ; Ke, Q ; Gong, M ; Drummond, T (IEEE COMPUTER SOC, 2023)
-
ItemNo Preview AvailableMaximum Spatial Perturbation Consistency for Unpaired Image-to-Image TranslationXu, Y ; Xie, S ; Wu, W ; Zhang, K ; Gong, M ; Batmanghelich, K (IEEE COMPUTER SOC, 2022)
-
ItemNo Preview AvailableAlleviating Semantics Distortion in Unsupervised Low-Level Image-to-Image Translation via Structure Consistency ConstraintGuo, J ; Li, J ; Fu, H ; Gong, M ; Zhang, K ; Tao, D (IEEE COMPUTER SOC, 2022)
-
ItemData-Driven Approach to Multiple-Source Domain AdaptationStojanov, P ; Gong, M ; Carbonell, J ; Zhang, K (PMLR, 2019)A key problem in domain adaptation is determining what to transfer across different domains. We propose a data-driven method to represent these changes across multiple source domains and perform unsupervised domain adaptation. We assume that the joint distributions follow a specific generating process and have a small number of identifiable changing parameters, and develop a data-driven method to identify the changing parameters by learning low-dimensional representations of the changing class-conditional distributions across multiple source domains. The learned low-dimensional representations enable us to reconstruct the target-domain joint distribution from unlabeled target-domain data, and further enable predicting the labels in the target domain. We demonstrate the efficacy of this method by conducting experiments on synthetic and real datasets.
-
ItemGeometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain MappingFu, H ; Gong, M ; Wang, C ; Batmanghelich, K ; Zhang, K ; Tao, D (IEEE, 2019)Unsupervised domain mapping aims to learn a function to translate domain X to Y by a function GXY in the absence of paired examples. Finding the optimal GXY without paired data is an ill-posed problem, so appropriate constraints are required to obtain reasonable solutions. One of the most prominent constraints is cycle consistency, which enforces the translated image by GXY to be translated back to the input image by an inverse mapping GYX. While cycle consistency requires the simultaneous training of GXY and GY X, recent studies have shown that one-sided domain mapping can be achieved by preserving pairwise distances between images. Although cycle consistency and distance preservation successfully constrain the solution space, they overlook the special properties that simple geometric transformations do not change the semantic structure of images. Based on this special property, we develop a geometry-consistent generative adversarial network (GcGAN), which enables one-sided unsupervised domain mapping. GcGAN takes the original image and its counterpart image transformed by a predefined geometric transformation as inputs and generates two images in the new domain coupled with the corresponding geometry-consistency constraint. The geometry-consistency constraint reduces the space of possible solutions while keep the correct solutions in the search space. Quantitative and qualitative comparisons with the baseline (GAN alone) and the state-of-the-art methods including CycleGAN and DistanceGAN demonstrate the effectiveness of our method.