Residual attention-based tracking-by-detection network with attention-driven data augmentation

Tracking-by-detection (TBD) is a significant framework for visual object tracking. However, current trackers are usually updated online based on random sampling with a probability distribution. The performance of the learning-based TBD trackers is limited by the lack of discriminative features, especially when the background is full of semantic distractors. We propose an attention-driven data augmentation method, in which a residual attention mechanism is integrated into the TBD tracking network as supplementary references to identify discriminative image features. A mask generating network is used to simulate changes in target appearances to obtain positive samples, where attention information and image features are combined to identify discriminative features. In addition, we propose a method for mining hard negative samples, which searches for semantic distractors with the response of the attention module. The experiments on the OTB2015, UAV123, and LaSOT benchmarks show that this method achieves competitive performance in terms of accuracy and robustness.     

Feedback-based object detection for multi-person pose estimation

In this paper, a novel method is proposed for increasing the performance through coupling of top-down models adjusting the object detector based on a new loss function. Generally, object detectors and keypoint estimators are sequentially used in real-time multi-person pose estimations; however, these two models are separately trained. Therefore, the results of the object detector are not optimized for the keypoint estimator. To solve this problem, we analyze the relationship between the two models and propose a feedback-based loss optimization in the object detector, based on the estimation results of the keypoint estimator. In addition, the resulting bounding box of the object detector is readjusted to improve the accuracy of the keypoint estimation model. The experimental results demonstrate that the proposed approach can perform real-time operations with a high frame rate similar to that of the baseline model. Moreover, it achieved an accuracy of 74.2 average precision (AP), which is higher than the state-of-the-arts model including the human detector used in the experiment.     

Quantum computation based bundling optimization for combinatorial auction in freight service procurements

Combinatorial auction is a useful trade manner for transportation service procurements in e-marketplaces. To enhance the competition of combinatorial auction, a novel auction mechanism of two-round bidding with bundling optimization is proposed. As the recommended the auction mechanism, the shipper/auctioneer integrates the objects into several bundles based on the bidding results of first round auction. Then, carriers/bidders bid for the object bundles in second round. The bundling optimization is described as a multi-objective model with two criteria on price complementation and combination consistency. A Quantum Evolutionary Algorithm (QEA) with β-based rotation gate and the encoding scheme based on non-zero elements in complementary coefficient matrix is developed for the model solution. Comparing with a Contrast Genetic Algorithm, QEA can achieve better computational performances for small and middle size problems.     

Improved non-maximum suppression for object detection using harmony search algorithm

Non-maximum suppression (NMS) plays a key role in many modern object detectors. It is responsible to remove detection boxes that cover the same object. NMS greedily selects the detection box with maximum score; other detection boxes are suppressed when the degree of overlap between these detection boxes and the selected box exceeds a predefined threshold. Such a strategy easily retain some false positives, and it limits the ability of NMS to perceive nearby objects in cluttered scenes. This paper proposes an effective method combining harmony search algorithm and NMS to alleviate this problem. This method regards the task of NMS as a combination optimization problem. It seeks final detection boxes under the guidance of an objective function. NMS is applied to each harmony to remove imprecise detection boxes, and the remaining boxes are used to calculate the fitness value. The remaining detection boxes in a harmony with highest fitness value are chosen as the final detection results. The standard Pattern Analysis, Statistical Modeling and Computational Learning Visual Object Classes dataset and the Microsoft Common Objects in Context dataset are used in all of the experiments. The proposed method is applied to two popular detection networks, namely Faster Region-based Convolutional Neural Networks and Region-based Fully Convolutional Networks. The experimental results show that the proposed method improves the average precision of these two detection networks. Moreover, the location performance and average recall of these two detectors are also improved.     

Using a Discrete-Event System Specifications (DEVS) for designing a Modelica compiler

We introduce a new architecture for the design of a tool for modeling and simulation of continuous and hybrid systems. The environment includes a compiler based on Modelica, a modular and a causal standard specification language for physical systems modeling (the tool supports models composed using certain component classes defined in the Modelica Standard Library, and the instantiation, parameterization and connection of these MSL components are described using a subset of Modelica). Models are defined in Modelica and are translated into DEVS models. DEVS theory (originally defined for modeling and simulation of discrete event systems) was extended in order to permit defining these of models. The different steps in the compiling process are show, including how to model these dynamic systems under the discrete event abstraction, including examples of model simulation with their execution results.     

工程教育认证背景下设计模式课程教学研究

工程教育是我国高等教育的重要组成部分,工程教育认证已成为中国高校新的风向标,也成为高考生选择报考专业的一项重要指标。设计模式是计算机科学与技术专业的一门核心课程,包含了面向对象的精髓,是软件体系结构的基础,在工程教育认证所评估的课程设置中起着承上启下的关键作用。本文基于工程教育专业认证理念,以及山东农业大学申请认证所用的《计算机科学与技术专业培养方案》,从教学内容、教学方式、课程目标与毕业要求对应关系、解决复杂工程问题能力、教学进度安排等方面,对设计模式课程开展了教学研究,并将该教学方案应用于本科教学。     

Bi-projection for 360°image object detection bridged by RoI Searcher

Object detection on 360°images is a vital component of 3D environment perception. The existing methods either treat panoramic images (usually represented in equirectangular projection—ERP) as normal FoV images and endure the distortions or project them into the less-distortion format and narrow the FoV, leading to unsatisfactory performance in practical applications. To solve this problem, we propose a dual-projection 360°object detection network named Bip R-CNN, consisting of three modules: a bi-projection feature extractor, a cross-projection region-of-interest (RoI) searcher, and a classification and regression predictor. Specifically, we extract the equirectangular and corresponding dual-cubemap features simultaneously from the input images. Besides, Projection-Inter Feature Fusion and Projection-Intra Feature Fusion are designed to allow the mutual interaction between the bi-projective features and promote the integration of features at different scales, respectively. In the proposed cross-projection RoI Searcher, we search for the bounding box (BBox) locations on cubemap from the corresponding ERP spherical proposals, bridging the RoIs of two different projection formats at feature level. Finally, the cube proposals are used to detect objects in the last predictor module. Considering the scarceness of the existing panoramic dataset (only indoor scenes), we propose an efficient approach to convert conventional datasets into annotated panoramic datasets without manual intervention, increasing the diversity of panoramic datasets. Extensive experiments are conducted on the synthetic and real-world datasets with spherical criteria, demonstrating our superiority to other state-of-the-art solutions.     

Unsupervised regions based segmentation using object discovery

We present a new unsupervised algorithm to discovery and segment out common objects from multiple images. Compared with previous cosegmentation methods, our algorithm performs well even when the appearance variations in the foregrounds are more substantial than those in some areas of the backgrounds. Our algorithm mainly includes two parts: the foreground object discovery scheme and the iterative region allocation algorithm. Two terms, a region-saliency prior and a region-repeatness measure, are introduced in the foreground object discovery scheme to detect the foregrounds without any supervisory information. The iterative region allocation algorithm searches the optimal solution for the final segmentation with the constraints from a maximal spanning tree, and an effective color-based model is utilized during this process. The comparative experimental results show that the proposed algorithm matches or outperforms several previous methods on several standard datasets.