Geng Sun

Jingru Zhu, Ya Guo, Geng Sun et al.

Semantic segmentation is a key technique involved in automatic interpretation of high-resolution remote sensing (HRS) imagery and has drawn much attention in the remote sensing community. Deep convolutional neural networks (DCNNs) have been successfully applied to the HRS imagery semantic segmentation task due to their hierarchical representation ability. However, the heavy dependency on a large number of training data with dense annotation and the sensitiveness to the variation of data distribution severely restrict the potential application of DCNNs for the semantic segmentation of HRS imagery. This study proposes a novel unsupervised domain adaptation semantic segmentation network (MemoryAdaptNet) for the semantic segmentation of HRS imagery. MemoryAdaptNet constructs an output space adversarial learning scheme to bridge the domain distribution discrepancy between source domain and target domain and to narrow the influence of domain shift. Specifically, we embed an invariant feature memory module to store invariant domain-level context information because the features obtained from adversarial learning only tend to represent the variant feature of current limited inputs. This module is integrated by a category attention-driven invariant domain-level context aggregation module to current pseudo invariant feature for further augmenting the pixel representations. An entropy-based pseudo label filtering strategy is used to update the memory module with high-confident pseudo invariant feature of current target images. Extensive experiments under three cross-domain tasks indicate that our proposed MemoryAdaptNet is remarkably superior to the state-of-the-art methods.

Xiaohua Ding, Hongyu Liu, Zaijiu Shang et al.

Based on reasonable testing model problems, we study the preservation by symplectic Runge-Kutta method (SRK) and symplectic partitioned Runge-Kutta method (SPRK) of structures for fixed points of linear Hamiltonian systems. The structure-preservation region provides a practical criterion for choosing step-size in symplectic computation. Examples are given to justify the investigation.

Geng Sun, Siqing Gan, Hongyu Liu et al.

Symmetric method and symplectic method are classical notions in the theory of Runge-Kutta methods. They can generate numerical flows that respectively preserve the symmetry and symplecticity of the continuous flows in the phase space. Adjoint method is an important way of constructing a new Runge-Kutta method via the symmetrisation of another Runge-Kutta method. In this paper, we introduce a new notion, called symplectic-adjoint Runge-Kutta method. We prove some interesting properties of the symmetric-adjoint and symplectic-adjoint methods. These properties reveal some intrinsic connections among several classical classes of Runge-Kutta methods. In particular, the newly introduced notion and the corresponding properties enable us to develop a novel and practical approach of constructing high-order explicit Runge-Kutta methods, which is a challenging and longly overlooked topic in the theory of Runge-Kutta methods.

Jingru Zhu, Ya Guo, Geng Sun et al.

Semantic segmentation of high-resolution remote sensing imagery (HRSI) suffers from the domain shift, resulting in poor performance of the model in another unseen domain. Unsupervised domain adaptive (UDA) semantic segmentation aims to adapt the semantic segmentation model trained on the labeled source domain to an unlabeled target domain. However, the existing UDA semantic segmentation models tend to align pixels or features based on statistical information related to labels in source and target domain data, and make predictions accordingly, which leads to uncertainty and fragility of prediction results. In this paper, we propose a causal prototype-inspired contrast adaptation (CPCA) method to explore the invariant causal mechanisms between different HRSIs domains and their semantic labels. It firstly disentangles causal features and bias features from the source and target domain images through a causal feature disentanglement module. Then, a causal prototypical contrast module is used to learn domain invariant causal features. To further de-correlate causal and bias features, a causal intervention module is introduced to intervene on the bias features to generate counterfactual unbiased samples. By forcing the causal features to meet the principles of separability, invariance and intervention, CPCA can simulate the causal factors of source and target domains, and make decisions on the target domain based on the causal features, which can observe improved generalization ability. Extensive experiments under three cross-domain tasks indicate that CPCA is remarkably superior to the state-of-the-art methods.

Xin-long Luo, Jia-hui Lv, Geng Sun

This paper investigates the localization problem of high-speed high-altitude unmanned aerial vehicle (UAV) with a monocular camera and inertial navigation system. It proposes a navigation method utilizing the complementarity of vision and inertial devices to overcome the singularity which arises from the horizontal flight of UAV. Furthermore, it modifies the mathematical model of localization problem via separating linear parts from nonlinear parts and replaces a nonlinear least-squares problem with a linearly equality-constrained optimization problem. In order to avoid the ill-condition property near the optimal point of sequential unconstrained minimization techniques(penalty methods), it constructs a semi-implicit continuous method with a trust-region technique based on a differential-algebraic dynamical system to solve the linearly equality-constrained optimization problem. It also analyzes the global convergence property of the semi-implicit continuous method in an infinity integrated interval other than the traditional convergence analysis of numerical methods for ordinary differential equations in a finite integrated interval. Finally, the promising numerical results are also presented.

Xin-Long Luo, Geng Sun

This article explores some geometric and algebraic properties of the dynamical system which is represented by matrix differential equations arising from inertial navigation problems, such as the symplecticity and the orthogonality. Furthermore, it extends the applicable fields of symplectic geometric algorithms from the even dimensional Hamiltonian system to the odd dimensional dynamical system. Finally, some numerical experiments are presented and illustrate the theoretical results of this paper.