The interpretability of the convolutional neural networks(CNNs) has become a research hotspot. A popular explanation method is based on Class Activation Mapping (CAM), which visualizes the salient regions most relevan...
The interpretability of the convolutional neural networks(CNNs) has become a research hotspot. A popular explanation method is based on Class Activation Mapping (CAM), which visualizes the salient regions most relevant to neural network decisions. However, many CAM methods use the feature maps produced by the final convolution layer to generate the class activation maps, which usually have a low spatial resolution and can only generate coarse-grained visual explanations that provide a rough spatial location of the target object. In this paper, we propose a novel CAM method named Fusion-CAM. It improves traditional CAM methods by combining final class activation map containing semantic information with intermediate layer class activation maps containing fine-grained details, to generate fine-grained visual explanations with high faithfulness. In order to obtain high-quality intermediate layer class activation maps, we utilize Layer-wise Relevance Propagation (LRP) to obtain the weighting components of each channel of the intermediate layer feature maps, and the intermediate layer class activation maps generated by weighted summation are less noisy and have clear fine-grained details, which help to improve the quality of the final class activation map. Qualitative and quantitative experiments show that Fusion-CAM can be easily attached to different CAM methods to improve their performance.
We present 3D Cinemagraphy, a new technique that marries 2D image animation with 3D photography. Given a single still image as input, our goal is to generate a video that contains both visual content animation and cam...
详细信息
ISBN:
(纸本)9798350301298
We present 3D Cinemagraphy, a new technique that marries 2D image animation with 3D photography. Given a single still image as input, our goal is to generate a video that contains both visual content animation and camera motion. We empirically find that naively combining existing 2D image animation and 3D photography methods leads to obvious artifacts or inconsistent animation. Our key insight is that representing and animating the scene in 3D space offers a natural solution to this task. To this end, we first convert the input image into feature-based layered depth images using predicted depth values, followed by unprojecting them to a feature point cloud. To animate the scene, we perform motion estimation and lift the 2D motion into the 3D scene flow. Finally, to resolve the problem of hole emergence as points move forward, we propose to bidirectionally displace the point cloud as per the scene flow and synthesize novel views by separately projecting them into target image planes and blending the results. Extensive experiments demonstrate the effectiveness of our method. A user study is also conducted to validate the compelling rendering results of our method.
Automatic image cropping algorithms aim to recompose images like human-being photographers by generating the cropping boxes with improved composition quality. Cropping box regression approaches learn the beauty of com...
详细信息
ISBN:
(纸本)9781577358800
Automatic image cropping algorithms aim to recompose images like human-being photographers by generating the cropping boxes with improved composition quality. Cropping box regression approaches learn the beauty of composition from annotated cropping boxes. However, the bias of annotations leads to quasi-trivial recomposing results, which has an obvious tendency to the average location of training samples. The crux of this predicament is that the task is naively treated as a box regression problem, where rare samples might be dominated by normal samples, and the composition patterns of rare samples are not well exploited. Observing that similar composition patterns tend to be shared by the cropping boundaries annotated nearly, we argue to find the beauty of composition from the rare samples by clustering the samples with similar cropping boundary annotations, i.e., similar composition patterns. We propose a novel Contrastive Composition Clustering (C2C) to regularize the composition features by contrasting dynamically established similar and dissimilar pairs. In this way, common composition patterns of multiple images can be better summarized, which especially benefits the rare samples and endows our model with better generalizability to render nontrivial results. Extensive experimental results show the superiority of our model compared with prior arts.
This article proposes a multi-agent deep reinforce-ment learning algorithm to control a fleet of unmanned surface vessels (USVs) that encircle and capture sea targets. First, a simulation environment for USVs is estab...
详细信息
Two goals of multi-objective evolutionary algorithms are effectively improving their convergence and diversity and making the Pareto set evenly distributed and close to the real Pareto front. At present, the challenge...
详细信息
Two goals of multi-objective evolutionary algorithms are effectively improving their convergence and diversity and making the Pareto set evenly distributed and close to the real Pareto front. At present, the challenges to be solved by the multi-objective evolutionary algorithm are to improve the convergence and diversity of the algorithm, and how to better solve functions with complex PF and/or PS shapes. Therefore, this paper proposes a gray wolf optimization-based self-organizing fuzzy multi-objective evolutionary algorithm. Gray wolf optimization algorithm is used to optimize the initial weights of the self-organizing map network. New neighborhood relationships for individuals are built by self-organizing map, which can maintain the invariance of feature distribution and map the structural information of the current population into Pareto sets. Based on this neighborhood relationship, this paper uses the fuzzy differential evolution operator, which constructs a fuzzy inference system to dynamically adjust the weighting parameter in the differential operator, to generate a new initial solution, and the polynomial mutation operator to refine them. Boundary processing is then conducted. Experiments on 15 problems of GLT1-6 and WFG1-9 and the algorithm proposed in this paper achieve the best on 18 values. And the result shows that the convergence and diversity of the proposed algorithm are better than several state-of-the-art multi-objective evolutionary algorithms.
In this paper, the problem of guarding a circular area is proposed and solved using a Stackelberg differential game theoretic approach. The objective of the attacker is to breach the perimeter of the defended area, wh...
详细信息
In this paper, the problem of guarding a circular area is proposed and solved using a Stackelberg differential game theoretic approach. The objective of the attacker is to breach the perimeter of the defended area, while the defenders endeavor to thwart such attempts. The dynamics of the attack-defense game are modeled according to the distance and position relations among defenders, attackers, and the center of defense area. The optimal Stackelberg equilibrium control strategies for both defenders and attackers are designed to guarantee the defense mission's success. Then, the effectiveness of the proposed method is validated through numerical simulation.
In this article, we design a distributed coordinated guiding vector field (CGVF) for a group of robots to achieve ordering-flexible motion coordination while maneuvering on a desired 2-D surface. The CGVF is character...
详细信息
In this article, we design a distributed coordinated guiding vector field (CGVF) for a group of robots to achieve ordering-flexible motion coordination while maneuvering on a desired 2-D surface. The CGVF is characterized by three terms, i.e., a convergence term to drive the robots to converge to the desired surface, a propagation term to provide a traversing direction for maneuvering on the desired surface, and a coordinated term to achieve the surface motion coordination with an arbitrary ordering of the robotic group. By setting the surface parameters as additional virtual coordinates, the proposed approach eliminates potential singularity of the CGVF and enables both the global convergence to the desired surface and the maneuvering on the surface from all possible initial conditions. The ordering-flexible surface motion coordination is realized by each robot to share with its neighbors only two virtual coordinates, i.e., that of a given target and that of its own, which reduces the communication and computation cost in multirobot surface navigation. Finally, the effectiveness of the CGVF is substantiated by extensive numerical simulations.
Current protein nuclear localization assays encounter multiple challenges that underscore the constraints of conventional biochemical assays and sequence-based procedures. This paper highlights the emerging interest i...
Current protein nuclear localization assays encounter multiple challenges that underscore the constraints of conventional biochemical assays and sequence-based procedures. This paper highlights the emerging interest in utilizing artificial intelligence to surmount these limitations. Specifically, a supervised deep learning algorithm, employing the HTRNN model, is presented to identify mechanisms responsible for the nuclear location of proteins. This methodology embraces an entire data-driven end-to-end structure, making expert experience or previous biological knowledge unnecessary. By utilizing amino acid sequence features from both the head and tail regions, the algorithm displays remarkable detection and generalization capabilities, which have been validated by experimental results on present tagged nuclear location protein datasets. Overall, the data-driven algorithm plays a crucial role in enhancing the identification of protein nuclear localization properties.
The microgrid (MG) system based on renewable energy generators plays a significant role in sustainable development and environmental protection, which has been developed rapidly. As a promising clean energy conversion...
详细信息
The microgrid (MG) system based on renewable energy generators plays a significant role in sustainable development and environmental protection, which has been developed rapidly. As a promising clean energy conversion technology, solid oxide fuel cell (SOFC) is a clean, efficient and controllable power generator, which is very suitable to be integrated in a distributed MG system. Finding the optimal size configuration and the beneficial operation strategy of the Grid-connected MG under different operation modes are critical issues for MG application. In this work, a single-dwelling MG incorporating solar photovoltaic (PV), wind turbine generator (WTG), SOFC and battery energy storage system (BESS) is studied by minimizing the system levelized cost of energy (LCOE) on the basis of system multi-constraints. The dispatching problem of MG is modeled as a quadratic programming problem and an improved GA-PSO algorithm is employed to explore the optimal configuration. Then, sensitivity analysis is carried out to identify the impact of different distributed energy resource size on the performance of MG. Based on the optimal configuration, the operation strategy of the proposed MG under both off-grid mode and grid-connected mode, as well as the influence of electricity price and fuel price on the operation of the MG are investigated. Finally, the economic and environmental benefits of the MG are studied and compared. The results show that, in Shanghai, Beijing, Wuhan and Hulunbuir, the costs of the MG in both off-grid mode and grid-connected mode are lower than the grid-supplied price by up to 13%-28% and 28%-37% separately.
The bending and twisting of DNA origami structures are important features for controlling the physical properties of DNA nanodevices. It has not been fully explored yet how to finely tune the bending and twisting of c...
详细信息
The bending and twisting of DNA origami structures are important features for controlling the physical properties of DNA nanodevices. It has not been fully explored yet how to finely tune the bending and twisting of curved DNA structures. Traditional tuning of the curved DNA structures was limited to controlling the in-plane-bending angle through varying the numbers of base pairs of deletions and insertions. Here, we developed two tuning strategies of curved DNA origami structures from in silico and in vitro aspects. In silico, the out-of-plane bending and twisting angles of curved structures were introduced, and were tuned through varying the patterns of base pair deletions and insertions. In vitro, a chemical adduct (ethidium bromide) was applied to dynamically tune a curved spiral. The 3D structural conformations, like chirality, of the curved DNA structures were finely tuned through these two strategies. The simulation and TEM results demonstrated that the patterns of base pair insertions and deletions and chemical adducts could effectively tune the bending and twisting of curved DNA origami structures. These strategies expand the programmable accuracy of curved DNA origami structures and have potential in building efficient dynamic functional nanodevices.
暂无评论