There are numerous applications for visual object tracking in computer vision, and it aims to attain the highest tracking reliability and accuracy depending on the applications' varied evaluation criteria. Althoug...
详细信息
There are numerous applications for visual object tracking in computer vision, and it aims to attain the highest tracking reliability and accuracy depending on the applications' varied evaluation criteria. Although DCF tracking algorithms have been used in the past and achieved great results, they are still unable to provide robust tracking under difficult conditions such as occlusion, scale fluctuation, quick motion, and motion blur. To address the instability during tracking brought on by various challenging issues in complex sequences, we present a novel framework termed improved spatial-temporal regularized correlation filters (I-STRCF) to integrate with instantaneous motion estimation and Kalman filter for visual object tracking which can minimize the possible tracking failure during tracking as the tracking model update itself with Kalman filter throughout the video sequence. We also include a unique scale estimate criterion called average peak-to-correlation energy to address the issue of target loss brought on by scale change. Using the previously calculated motion data, the suggested method predicts the potential scale region of the target in the current frame, and then the target model updates the target object's position in successive frames. Additionally, we examine the factors affecting how well the suggested framework performs in extensive experiments. The experimental results show that this proposed framework achieves the best visual tracking for computer vision and performs better than STRCF on Temple Color-128 datasets for object tracking attributes. Our framework produces greater AUC improvements for the scale variation, background clutter, lighting variation, occlusion, out-of-plane rotation, and deformation properties when compared to STRCF. Our system gets much better improvements than its rivals in terms of performance and robustness for sporting events.
The tracker based on correlation filters can achieve effective positioning at a relatively fast speed, resulting from its operation in the frequency domain. As a result, it is commonly employed in the field of object ...
详细信息
The tracker based on correlation filters can achieve effective positioning at a relatively fast speed, resulting from its operation in the frequency domain. As a result, it is commonly employed in the field of object tracking. However, this characteristic introduces boundary effect and affects the tracking performance in some scenes. In this work, a correlation filter tracking algorithm with spatial-temporal regularization and context awareness (STCACF) is proposed: (1) the spatial-temporal information and context awareness is added to the training process to mitigate the boundary effect and enhance the overall tracking performance;(2) the tracker model adopts the iterative method of alternating direction method of multipliers (ADMM), so that each subproblem can be solved in a closed-loop solution, which can realize real-time tracking;(3) the spatialregularization is employed to reduce the influence of filter degradation. Experiments on the OTB-2013, the OTB-2015 and the TC-128 benchmark datasets demonstrate that the suggested STCACF is capable of significantly improving the tracking performance compared with state-of-the-art trackers. The STCACF tracker runs at a frame rate of approximately 22 frames per second (FPS) on a single central processing unit (CPU).
Correlation filters are known to have superior performance in tracking speed. To improve tracking results, we propose a context-aware regression correlation filter with a spatial-temporal regularization for tracking. ...
详细信息
Correlation filters are known to have superior performance in tracking speed. To improve tracking results, we propose a context-aware regression correlation filter with a spatial-temporal regularization for tracking. First, the spatialregularization parameters are computed by the spatial correlation between the target and the surrounding information. Meanwhile, the filter weight distribution is also computed to highlight the target region and suppress the background region. Next, a context-aware model is proposed to adaptively expand the search area of the target with the original regression analysis. Due to the introduction of the context-aware information, the target structure will be changed and the regression model will not be adapted to the Gaussian function. Therefore, an optimized regression objective function is constructed according to the context-aware model. In the tracking process, under rotation, out-of-view, deformation, etc., the tracker still can continue the tracking of subsequent frames according to partial background information. In addition, time variables for online updating of the spatial-temporal model employ the similarity of adjacent frames to achieve more accurate tracking. According to the model form, the alternative direction method of multipliers is used to optimize the model optimization. Extensive experimental results on the OTB-2013 and OTB-2015 dataset object tracking benchmarks prove that the proposed algorithm is superior to other state-of-the-art algorithms in terms of success, accuracy, and robustness. (C) 2020 SPIE and IS&T
The infrared small target detection is crucial for the efficacy of infrared search and tracking systems. Current tensor decomposition methods emphasize representing small targets with sparsity but struggle to separate...
详细信息
The infrared small target detection is crucial for the efficacy of infrared search and tracking systems. Current tensor decomposition methods emphasize representing small targets with sparsity but struggle to separate targets from complex backgrounds due to insufficient use of intrinsic directional information and reduced target visibility during decomposition. To address these challenges, this study introduces a sparse differential directionality prior (SDD) framework. SDD leverages the distinct directional characteristics of targets to differentiate them from the background, applying mixed sparse constraints on the differential directional images and continuity difference matrix of the temporal component, both derived from Tucker decomposition. We further enhance the target detectability with a saliency coherence strategy that intensifies target contrast against the background during hierarchical decomposition. A proximal alternating minimization (PAM)-based algorithm efficiently solves our proposed model. Experimental results on several real-world datasets validate our method's effectiveness, outperforming ten state-of-the-art methods in target detection and clutter suppression. Our code is available at: https://***/GrokCV/SDD.
Despite excellent performance shown by spatially regularized discriminative correlation filters (SRDCF) for visual tracking, some issues remain open that hinder further boosting their performance: first, SRDCF utilize...
详细信息
Despite excellent performance shown by spatially regularized discriminative correlation filters (SRDCF) for visual tracking, some issues remain open that hinder further boosting their performance: first, SRDCF utilizes multiple training images to formulate its model, which makes it unable to exploit the circulant structure of the training samples in learning, leading to high computational burden;second, SRDCF is unable to efficiently exploit the powerfully discriminative nonlinear kernels, further negatively affecting its performance. In this paper, we present a novel spatial-temporally regularized complementary kernelized CFs (STRCKCF) based tracking approach. First, by introducing spatial-temporal regularization to the filter learning, the STRCKCF formulates its model with only one training image, which can not only facilitate exploiting the circulant structure in learning, but also reasonably approximate the SRDCF with multiple training images. Furthermore, by incorporating two types of kernels whose matrices are circulant, the STRCKCF is able to fully take advantage of the complementary traits of the color and HOG features to learn a robust target representation efficiently. Besides, our STRCKCF can be efficiently optimized via the alternating direction method of multipliers (ADMM). Extensive evaluations on OTB100 and VOT2016 visual tracking benchmarks demonstrate that the proposed method achieves favorable performance against state-of-the-art trackers with a speed of 40fpson a single CPU. Compared with SRDCF, STRCKCF provides a 8 x speedup and achieves a gain of 5.5% AUC score on OTB100 and 8.4% EAO score on VOT2016.
There are many cases where one needs to limit the X-ray dose, or the number of projections, or both, for high frame rate (fast) imaging. Normally, it improves temporal resolution but reduces the spatial resolution of ...
详细信息
There are many cases where one needs to limit the X-ray dose, or the number of projections, or both, for high frame rate (fast) imaging. Normally, it improves temporal resolution but reduces the spatial resolution of the reconstructed data. Fortunately, the redundancy of information in the temporal domain can be employed to improve spatial resolution. In this paper, we propose a novel regularizer for iterative reconstruction of time-lapse computed tomography. The non-local penalty term is driven by the available prior information and employs all available temporal data to improve the spatial resolution of each individual time frame. A high-resolution prior image from the same or a different imaging modality is used to enhance edges which remain stationary throughout the acquisition time while dynamic features tend to be regularized spatially. Effective computational performance together with robust improvement in spatial and temporal resolution makes the proposed method a competitive tool to state-of-the-art techniques.
暂无评论