Discriminative correlation filters (DCFs) have been shown to perform superiorly in visual tracking. They only need a small set of training samples from the initial frame to generate an appearance model. However, existing DCFs learn the filters separately from feature extraction, and update these filters using a moving average operation with an empirical weight. These DCF trackers hardly benefit from the end-to-end training. In this paper, we propose the CREST algorithm to reformulate DCFs as a one-layer convolutional neural network. Our method integrates feature extraction, response map generation as well as model update into the neural networks for an end-to-end training. To reduce model degradation during online update, we apply residual learning to take appearance changes into account. Extensive experiments on the benchmark datasets demonstrate that our CREST tracker performs favorably against state-of-the-art trackers. 1
The tracking-by-detection framework consists of two stages, i.e., drawing samples around the target object in the first stage and classifying each sample as the target object or as background in the second stage. The performance of existing trackers using deep classification networks is limited by two aspects. First, the positive samples in each frame are highly spatially overlapped, and they fail to capture rich appearance variations. Second, there exists extreme class imbalance between positive and negative samples. This paper presents the VITAL algorithm to address these two problems via adversarial learning. To augment positive samples, we use a generative network to randomly generate masks, which are applied to adaptively dropout input features to capture a variety of appearance changes. With the use of adversarial learning, our network identifies the mask that maintains the most robust features of the target objects over a long temporal span. In addition, to handle the issue of class imbalance, we propose a highorder cost sensitive loss to decrease the effect of easy negative samples to facilitate training the classification network. Extensive experiments on benchmark datasets demonstrate that the proposed tracker performs favorably against stateof-the-art approaches.
We propose an unsupervised visual tracking method in this paper. Different from existing approaches using extensive annotated data for supervised learning, our CNN model is trained on large-scale unlabeled videos in an unsupervised manner. Our motivation is that a robust tracker should be effective in both the forward and backward predictions (i.e., the tracker can forward localize the target object in successive frames and backtrace to its initial position in the first frame). We build our framework on a Siamese correlation filter network, which is trained using unlabeled raw videos. Meanwhile, we propose a multiple-frame validation method and a cost-sensitive loss to facilitate unsupervised learning. Without bells and whistles, the proposed unsupervised tracker achieves the baseline accuracy of fully supervised trackers, which require complete and accurate labels during training. Furthermore, unsupervised framework exhibits a potential in leveraging unlabeled or weakly labeled data to further improve the tracking accuracy. * Y. Song and W. Liu are the corresponding authors. This work is done when N. Wang is an intern in Tencent AI Lab. The source code and results are available at https://github.com/594422814/UDT. Supervised Training: Annotated sequences Forward tracking Unsupervised Training: Unlabeled sequences Forward and Backward tracking
which has caused serious global warming. According to Intergovernmental Panel on Climate Change, global warming of 1.5 °C above preindustrial levels will come true by 2055 and bring about severe environmental and ecological issues, [2] such as ocean acidification, sea level rise, and species extinction. Therefore, it is urgently needed to reduce the atmospheric CO 2 concentration. [3] Apart from utilizing renewable energy resources to substitute for fossil fuels, CO 2 capture and storage (CCS) processes, and CO 2 capture and utilization (CCU) processes have been developed to effectively diminish the existing atmospheric CO 2 concentration. [4] The CCS processes are often expensive and laborious, and face the risk of CO 2 leakage, while the CCU processes are able to convert the captured CO 2 to value-added carbon-containing chemicals powered by external energy and has attracted more and more research interests recently. However, the conversion of CO 2 to target products is relatively difficult due to its extremely stable CO bonds. Several approaches for CO 2 conversion have been developed, such as photosynthesis, thermocatalytic reduction, electrochemical reduction, and photochemical conversion. Compared with other approaches, electrochemical reduction is more attractive because of two reasons: 1) electrochemical reduction process is controllable through adjusting the applied voltage and reaction temperature, and 2) the process can utilize renewable energy resources such as wind, solar, hydroelectric, and geothermal energies to electrochemically convert CO 2 to useful chemicals, which forms a carbon-neutral energy cycle and simultaneously provides an efficient storage method for renewable energy resources.Several types of electrolysis cell have been systematically investigated for CO 2 electroreduction as listed in Table 1. The first type operates in H-shaped electrochemical cells with liquid electrolyte at low temperatures (<100 °C).[5] Gaseous CO 2 reactant is dissolved into the liquid electrolyte and transported to the electrode surface, which is subsequently electroreduced to CO, HCOOH, CH 4 , C 2 H 4 , C 2 H 5 OH, and so on. [1,6] Although high Faradaic efficiency of products from CO 2 electroreduction has been achieved by exploring efficient catalysts recently, the current density of CO 2 electrolysis is relatively low due to the low CO 2 solubility and the competitive High-temperature CO 2 electrolysis in solid-oxide electrolysis cells (SOECs) could greatly assist in the reduction of CO 2 emissions by electrochemically converting CO 2 to valuable fuels through effective electrothermal activation of the stable CO bond. If powered by renewable energy resources, it could also provide an advanced energy-storage method for their intermittent output. Compared to low-temperature electrochemical CO 2 reduction, CO 2 electrolysis in SOECs at high temperature exhibits higher current density and energy efficiency and has thus attracted much recent attention. The history of its development and its fundamental mech...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.