Kernelized Correlation Filter Research Articles

Anti-Occlusion Visual Tracking Algorithm for UAVs with Multi-Feature Adaptive Fusion

Most of the existing trackers based on discriminative correlation filters use only one feature or a simple linear fusion of multiple features for object tracking, and most of them lack a mechanism to handle occlusions. This leads to poor tracking performance in rapidly changing and easily occluded scenarios, especially on unmanned aerial vehicle (UAV) platforms. To address this issue, this paper proposes an anti-occlusion visual tracking algorithm for UAVs with multi-feature adaptive fusion named multi-feature adaptive fusion and anti-occlusion tracker (MAFAOT). It introduces a novel approach for implementing an adaptive fusion of multiple features. This method transforms the multi-feature fusion problem into a maximization issue by designing a tracking quality evaluation index. It successfully achieves an adaptive fusion of gradient direction histogram and color histogram feature responses. MAFAOT also introduces an anti-occlusion update pool strategy, enabling the tracker to adapt dynamically to various complex scenarios, including occlusion and motion blur. The experimental results on the OTB100 and UAV123 datasets confirm the significant advantages of MAFAOT in terms of precision and success rate compared to other correlation filter-based algorithms. The proposed methods further enhance the expressiveness of the features and effectively avoid the problem of tracker contamination caused by occlusion. Furthermore, this paper applies the proposed methods to the kernelized correlation filters (KCF) algorithm. On the OTB100 dataset, the improved KCF algorithm shows an improvement of 10.94% in precision and 11.11% in success rate. On the UAV123 dataset, it shows an improvement of 14.53% in precision and 16.62% in success rate, further verifying the effectiveness and versatility of the proposed methods.

Design and Testing of an Autonomous Navigation Unmanned Surface Vehicle for Buoy Inspection

In response to the inefficiencies and high costs associated with manual buoy inspection, this paper presents the design and testing of an Autonomous Navigation Unmanned Surface Vehicle (USV) tailored for this purpose. The research is structured into three main components: Firstly, the hardware framework and communication system of the USV are detailed, incorporating the Robot Operating System (ROS) and additional nodes to meet practical requirements. Furthermore, a buoy tracking system utilizing the Kernelized Correlation Filter (KCF) algorithm is introduced. Secondly, buoy image training is conducted using the YOLOv7 object detection algorithm, establishing a robust model for accurate buoy state recognition. Finally, an improved Line-of-Sight (LOS) method for USV path tracking, assuming the presence of an attraction potential field around the inspected buoy, is proposed to enable a comprehensive 360-degree inspection. Experimental testing includes validation of buoy image target tracking and detection, assessment of USV autonomous navigation and obstacle avoidance capabilities, and evaluation of the enhanced LOS path tracking algorithm. The results demonstrate the USV's efficacy in conducting practical buoy inspection missions. This research contributes insights and advancements to the fields of maritime patrol and routine buoy inspections.

A modular motion compensation pipeline for prospective respiratory motion correction of multi-nuclear MR spectroscopy

Magnetic resonance (MR) acquisitions of the torso are frequently affected by respiratory motion with detrimental effects on signal quality. The motion of organs inside the body is typically decoupled from surface motion and is best captured using rapid MR imaging (MRI). We propose a pipeline for prospective motion correction of the target organ using MR image navigators providing absolute motion estimates in millimeters. Our method is designed to feature multi-nuclear interleaving for non-proton MR acquisitions and to tolerate local transmit coils with inhomogeneous field and sensitivity distributions. OpenCV object tracking was introduced for rapid estimation of in-plane displacements in 2D MR images. A full three-dimensional translation vector was derived by combining displacements from slices of multiple and arbitrary orientations. The pipeline was implemented on 3 T and 7 T MR scanners and tested in phantoms and volunteers. Fast motion handling was achieved with low-resolution 2D MR image navigators and direct implementation of OpenCV into the MR scanner’s reconstruction pipeline. Motion-phantom measurements demonstrate high tracking precision and accuracy with minor processing latency. The feasibility of the pipeline for reliable in-vivo motion extraction was shown on heart and kidney data. Organ motion was manually assessed by independent operators to quantify tracking performance. Object tracking performed convincingly on 7774 navigator images from phantom scans and different organs in volunteers. In particular the kernelized correlation filter (KCF) achieved similar accuracy (74%) as scored from inter-operator comparison (82%) while processing at a rate of over 100 frames per second. We conclude that fast 2D MR navigator images and computer vision object tracking can be used for accurate and rapid prospective motion correction. This and the modular structure of the pipeline allows for the proposed method to be used in imaging of moving organs and in challenging applications like cardiac magnetic resonance spectroscopy (MRS) or magnetic resonance imaging (MRI) guided radiotherapy.

Zynq FPGA-Based Acceleration of Kernelized Correlation Filters via High-Level Synthesis of a Custom DFT Block

This study presents a hardware-software co-design implementation of an accelerator for the Kernelized Correlation Filter (KCF) tracking algorithm. Leveraging the High-level synthesis (HLS) and the Zynq heterogeneous platform, the KCF algorithm’s performance is enhanced by using a custom hardware implementation for the computationally intensive Discrete Fourier Transform (DFT) operation. Within this framework, a custom combined DFT and inverse DFT IP, named CDFT, is developed and optimized on the Programmable Logic (PL) side of the Xilinx ZCU102 FPGA, whereas the rest of the KCF algorithm is run with customized Petalinux build on the (Processing System) side. To assess real-world performance, a driver for the CDFT IP and a user application were created to measure metrics like Center Location Error (CLE), Intersection over Union (IoU), and Frame per Second (FPS). The designed DFT accelerator achieves a remarkable speedup of 21x compared to a software DFT implementation. At the algorithm level, the KCF accelerator obtains a 6x speed up with negligible precision loss. In comparison to prior studies employing exclusively hardware implementations, the proposed approach demonstrates a high accuracy at a moderate speed, while there exists potential for further optimizations to enhance its performance even further.

Sign Language Detection using ML Technologies

Sign language recognition is a leap forward for helping deaf-mute people. In sign language, every sign has a meaning assigned to it, so that it becomes easy to understand and interpret by the people. The main primary objective of our project is to bridge the gap between the deaf and dumb people and ordinary people without the need of an intermediary translator. Gesture recognition is a computer based visual technique used to detect and identify the object in an image or video. An application of this gesture recognition technique involves translating the sign language into the American language which can be further understood and interpreted by normal people. Many researchers have proposed their systems for the implementation of the ASL system. This report is a review of some studies related to the same topic. We addressed different approaches that used Convolutional Neural Networks (CNNs), K- Nearest Neighbours algorithm (KNN), Edge detection algorithm, Deep Neural Network (DNN), fuzzy clustering machine learning algorithm, Kernelized correlation filters (KCF) algorithm along with their results and drawbacks. Furthermore, we report on research gaps while summarizing these studies. From the above research papers, we got an average accuracy of 90.24%. To overcome the drawbacks, we propose a system which collects images for deep learning using webcam and OpenCV, with the help of TensorFlow Object Detection and Python that allows you to translate sign language in real time

Object tracking method based on edge detection and morphology

With the continuous development of science and technology, intelligent surveillance technology using image processing and computer vision is also progressing. To improve the performance of target detection and tracking, an improved target tracking method is proposed, which uses a combination of the Canny operator and morphology for the detection part, and a Kalman filter extended Kernel Correlation Filter (KCF) tracking algorithm approach for the tracking part. First, a convolution kernel of 3×3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$3\ imes 3$$\\end{document} is improved to a convolution kernel of 2×2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$2\ imes 2$$\\end{document} in the traditional Canny algorithm, and the pixel gradient in the diagonal direction is increased. Secondly, a mathematical morphology theory of nonlinear filtering is applied to the Canny edge detection algorithm, and this method effectively improves the clarity of image edges. Finally, the extended kernel correlation filtering algorithm is applied to video surveillance and Online Object Tracking Benckmark2013 (OTB2013) datasets for testing. The experimental results show that the method proposed in this paper can accurately detect moving targets and the algorithm has good accuracy and success rate.

An adaptive kernel correlation filter algorithm and synchronization error correction method for 3D motion reconstruction of flexible flapping wings using binocular vision

In order to measure the kinematic parameters and deformation of flapping wings precisely, this paper presents a wing motion reconstruction method utilizing a binocular vision system (BVS) comprising two action cameras. We designed an Adaptive Kernel Correlation Filter algorithm based on target scaling prediction for wing reconstruction to improve target tracking accuracy and robustness. A binocular error correction method with a startup time difference-induced error model was proposed, which helps the action cameras achieve the same high-precision synchronization as the industrial cameras with external trigger mode. Moreover, an infrared motion capture system is applied to assess and compensate for errors in the BVS. The experimental results show that the maximum value of the average errors of the BVS is 1.352 mm, and the minimum similarity value of the two wings is 0.996. The results contribute to the deformation measurement and optimization of soft materials such as the flapping wings.

Hyperspectral low altitude UAV target tracking algorithm based on deep learning and improved KCF

This article presents a novel target tracking algorithm for hyperspectral low altitude UAV, combining deep learning with an improved Kernelized Correlation Filter (KCF). Initially, an image noise reduction method based on principal component analysis with Block-Matching 3D (BM3D), is employed to process redundant information. Subsequently, an image fusion method is utilized to merge the processed hyperspectral image and the high-resolution panchromatic band image to obtain a high spatial resolution image for target enhancement. Following this, YOLOv5 is used to detect the coordinate information of the UAV target in the current frame. Then, The KCF algorithm is used for target tracking in the current frame using kernel correlation filtering. Finally, the Discriminative Scale Spatial Tracker (DSST) is employed to determine the scale information to achieve a multi-scale tracking effect. The experimental results demonstrate that the algorithm presented in this paper surpasses CSK, HLT, and the conventional KCF algorithm in hyperspectral UAV datasets. On average, there is a significant increase in accuracy which is over 17% when using our algorithm.

Unified seam tracking algorithm via three-point weld representation for autonomous robotic welding

Autonomous robotic welding based on real-time seam tracking has shown excellent prospects in the field of intelligent manufacturing. In previous research, real-time seam tracking requires teaching trajectories, which has significant limitations. Moreover, as different types of welds have different shapes, their geometric distribution also exists in various forms such as straight lines, planar curves, and spatial curves, which brings significant challenges to the visual recognition of welds. The use of existing methods for tracking specific types of welds has limited applications. Currently, the existing methods are carried out for tracking specific types of welds, resulting in limited application scope. To handle various types of weld joints, a unified tracking paradigm named the three-point seam tracking algorithm (TSTA) is proposed in this study. First, a feature description method using three feature points is established to define different weld joints. The feature points are employed to reconstruct the position and attitude of welds, which allows the method to track welds of arbitrary geometries. Subsequently, a recognition method for the feature points combining morphology extraction and kernel correlation filter (KCF) tracker is designed. Combined with their respective advantages, the algorithm can identify the three feature points accurately and robustly in strong-noise environments. Extensive experiments including actual welding were conducted to verify the superiority of the proposed TSTA. Eleven representative types of weld seams were tested and all of them could be tracked in real time without any prior information. Compared with similar studies, the proposed recognition algorithm currently achieves the highest accuracy and best robustness. The TSTA provides a complete solution in autonomous robotic welding for unknown welds and shows excellent application prospects.

Behavioral response of fish under ammonia nitrogen stress based on machine vision

The long-term accumulation of ammonia nitrogen in aquaculture seriously affects the life of fish and even causes large-scale death. Moreover, when the concentration of ammonia nitrogen starts to accumulate, it is a judgment standard to provide early warning through the changes in fish behavior to prevent excessive ammonia nitrogen in water. Therefore, this paper proposes a novel approach to monitoring water quality for aquaculture based on deep learning and three-dimensional movement trajectory. The improved YOLOv8 model was used as the object detection approach to obtain three-dimensional position information of fish by combining Kalman filter, Kuhn Munkres (KM) algorithm, and Kernelized Correlation Filters (KCF) algorithm. The proposed approach was evaluated in the recovery experiment of acute ammonia nitrogen stress of sturgeon, bass, and crucian. The experimental results show that the precision, recall, mAP@0.5, and mAP@0.5:0.95 of the improved YOLOv8 model are 0.964, 0.914, 0.979, and 0.602, respectively. In addition, the proposed three-dimensional positioning approach can qualitatively and quantitatively analyze the fish behavior in different stages and further explores the fish behavior changes through behavior trajectories, volumes of exercise, spatial distribution, and movement velocity. This research provides a new method and idea for studying the abnormal behavior of aquatic animals under ammonia nitrogen stress and has theoretical and practical significance.

A target tracking method based on adaptive occlusion judgment and model updating strategy

Target tracking is an important research in the field of computer vision. Despite the rapid development of technology, difficulties still remain in balancing the overall performance for target occlusion, motion blur, etc. To address the above issue, we propose an improved kernel correlation filter tracking algorithm with adaptive occlusion judgement and model updating strategy (called Aojmus) to achieve robust target tracking. Firstly, the algorithm fuses color-naming (CN) and histogram of gradients (HOG) features as a feature extraction scheme and introduces a scale filter to estimate the target scale, which reduces tracking error caused by the variations of target features and scales. Secondly, the Aojmus introduces four evaluation indicators and a double thresholding mechanism to determine whether the target is occluded and the degree of occlusion respectively. The four evaluation results are weighted and fused to a final value. Finally, the updating strategy of the model is adaptively adjusted based on the weighted fusion value and the result of the scale estimation. Experimental evaluations on the OTB-2015 dataset are conducted to compare the performance of the Aojmus algorithm with four other comparable algorithms in terms of tracking precision, success rate, and speed. The experimental results show that the proposed Aojmus algorithm outperforms all the algorithms compared in terms of tracking precision. The Aojmus also exhibits excellent performance on attributes such as target occlusion and motion blur in terms of success rate. In addition, the processing speed reaches 74.85 fps, which also demonstrates good real-time performance.

Multi-Object Tracking Based on a Novel Feature Image With Multi-Modal Information

Multi-object tracking technology plays a crucial role in many applications, such as autonomous vehicles and security monitoring. This paper proposes a multi-object tracking framework based on the multi-modal information of 3D point clouds and color images. At each sampling instant, the 3D point cloud and image acquired by a LiDAR and a camera are fused into a color point cloud, where objects are detected by the Point-GNN method. And, a novel height-intensity-density (HID) image is constructed from the bird's eye view. The HID image truly reflects the shapes and materials of objects and effectively avoids the influence of object occlusion, which is helpful to object tracking. In two sequential HID images, a new rotation kernel correlation filter is proposed to predict the objects. Furthermore, an object retention module and an object re-recognition module are developed to overcome the object matching failure in the in-between frames. The proposed method takes full advantage of the multi-modal data and effectively achieves the information complementation to improve the accuracy of multi-object tracking. The experiments with the KITTI dataset show that the proposed method has the best performance among the existing traditional multi-object tracking methods.

Anti-Similar Visual Target Tracking Algorithm Based on Filter Peak Guidance and Fusion Network

Visual tracking is a key research area in computer vision, as tracking technology is increasingly being applied in daily life, it has high-research significance. Visual tracking technology usually faces various challenging interference factors, among which, a similar background is one of the factors that has a greater impact on the tracking process. Kernelized Correlation Filter (KCF) tracking algorithm can track targets quickly by using circulant matrix, and has good tracking effect, so it is widely used in the tracking field. However, when the target is interfered by similar objects, the filter template in KCF cannot effectively distinguish between the target and the interfering object. This is because the filter only uses the texture gradient feature as the description object of the target, which will make the KCF algorithm extremely sensitive to the change of the target; therefore, the filter has difficultly making a judgment in the unstable scene, cannot accurately describe the target state, and finally leads to tracking failure. Therefore, this paper fuses Color Names (CN) on the basis of the original Histogram of Oriented Gradients (HOG) feature of KCF, which can obtain a more comprehensive feature representation, and realize the application of combined features to improve the anti-interference ability of KCF in complex scenes. In addition, this paper also uses the peak response of correlation filtering as the judgment condition to determine whether the current tracking result is stable. When the filter is in an unstable tracking state, the proposed algorithm will select the value with high confidence from its multiple responses as the candidate target of the Siamese network, and the deep learning network is used as the incremental learning method of the filter. The Channel Attention is introduced into the network layer, so that the network can adaptively reason and adjust the extracted universal features, and the enhanced feature information is used as the final discriminant basis. Finally, according to the response, the target with the smallest error compared with the target template is selected from multiple candidate targets as the final tracking result. The experimental results show that the average accuracy and average success rate of the proposed algorithm are significantly improved compared with the classical tracking algorithm, especially in dealing with similar target interference.

Pedestrian Tracking in UAV Images With Kalman Filter Motion Estimator and Correlation Filter

Visual object tracking is one of the significant parts of systems varying from autonomous driving to drone-based surveillance/tracking. Discriminative correlation filter (DCF) based trackers have proved their prominence in the past few years. These utilize visual information present in images for monitoring. This work uses the Kalman Filter (KF) to derive a motion estimation model. It combines it with two DCF-based trackers, namely Kernelized Correlation Filter (KCF) and Discriminative Correlation Filter with Channel and Spatial Reliability (CSRDCF) for pedestrian tracking. Camera motion-compensated versions of the trackers are also presented. The performance of the proposed methodology is presented in terms of Success Rate (SR) and Precision (P). Real-time power consumption, memory occupancy, and speed of trackers on the Jetson Nano (ARM Cortex-A57, 4 GB RAM) board have been presented.

Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection

Hand detection and tracking are key components in many computer vision applications, including hand pose estimation and gesture recognition for human–computer interaction systems, virtual reality, and augmented reality. Despite their importance, reliable hand detection in cluttered scenes remains a challenge. This study explores the use of deep learning techniques for fast and robust hand detection and tracking. A novel algorithm is proposed by combining the Kernelized Correlation Filter (KCF) tracker with the Single-Shot Detection (SSD) method. This integration enables the detection and tracking of hands in challenging environments, such as cluttered backgrounds and occlusions. The SSD algorithm helps reinitialize the KCF tracker when it fails or encounters drift issues due to sudden changes in hand gestures or fast movements. Testing in challenging scenes showed that the proposed tracker achieved a tracking rate of over 90% and a speed of 17 frames per second (FPS). Comparison with the KCF tracker on 17 video sequences revealed an average improvement of 13.31% in tracking detection rate (TRDR) and 27.04% in object detection error (OTE). Additional comparison with MediaPipe hand tracker on 10 hand gesture videos taken from the Intelligent Biometric Group Hand Tracking (IBGHT) dataset showed that the proposed method outperformed the MediaPipe hand tracker in terms of overall TRDR and tracking speed. The results demonstrate the promising potential of the proposed method for long-sequence tracking stability, reducing drift issues, and improving tracking performance during occlusions.

Multi-objective tracking for smart substation onsite surveillance based on YOLO Approach and AKCF

The onsite surveillance plays an important role in the smart substation since the smart substation is unattended. All the sites and operation staff should be supervised throughout the process since a series of risks exist on the working sites. KCF (Kernel Correlation Filter) is an effective method to track a moving object for safety surveillance. However, the occlusion and shape changes worsen the performance of KCF, especially on the occasion of multi-objective detection. This paper proposes a comprehensive method for improving the precision and robustness of detection. Firstly, all the moving objects are detected by the YOLO method. In the tracking part, an AKCF (Augmented Kernel Correlation Filter) is proposed for the heavily occluded object, and the Kalman Filter (KF) serves as a supplementary output. Moreover, in the target association section, based on priority matching and rematching based on motion estimation, a two-stage target association method is proposed. Test outcomes indicate that the proposed algorithm is accurate and robust for tracking workers’ trajectories and conducting surveillance.

An Automated Skill Assessment Framework Based on Visual Motion Signals and a Deep Neural Network in Robot-Assisted Minimally Invasive Surgery.

Surgical skill assessment can quantify the quality of the surgical operation via the motion state of the surgical instrument tip (SIT), which is considered one of the effective primary means by which to improve the accuracy of surgical operation. Traditional methods have displayed promising results in skill assessment. However, this success is predicated on the SIT sensors, making these approaches impractical when employing the minimally invasive surgical robot with such a tiny end size. To address the assessment issue regarding the operation quality of robot-assisted minimally invasive surgery (RAMIS), this paper proposes a new automatic framework for assessing surgical skills based on visual motion tracking and deep learning. The new method innovatively combines vision and kinematics. The kernel correlation filter (KCF) is introduced in order to obtain the key motion signals of the SIT and classify them by using the residual neural network (ResNet), realizing automated skill assessment in RAMIS. To verify its effectiveness and accuracy, the proposed method is applied to the public minimally invasive surgical robot dataset, the JIGSAWS. The results show that the method based on visual motion tracking technology and a deep neural network model can effectively and accurately assess the skill of robot-assisted surgery in near real-time. In a fairly short computational processing time of 3 to 5 s, the average accuracy of the assessment method is 92.04% and 84.80% in distinguishing two and three skill levels. This study makes an important contribution to the safe and high-quality development of RAMIS.

Automatic identification and analysis of multi-object cattle rumination based on computer vision.

Rumination in cattle is closely related to their health, which makes the automatic monitoring of rumination an important part of smart pasture operations. However, manual monitoring of cattle rumination is laborious and wearable sensors are often harmful to animals. Thus, we propose a computer vision-based method to automatically identify multi-object cattle rumination, and to calculate the rumination time and number of chews for each cow. The heads of the cattle in the video were initially tracked with a multi-object tracking algorithm, which combined the You Only Look Once (YOLO) algorithm with the kernelized correlation filter (KCF). Images of the head of each cow were saved at a fixed size, and numbered. Then, a rumination recognition algorithm was constructed with parameters obtained using the frame difference method, and rumination time and number of chews were calculated. The rumination recognition algorithm was used to analyze the head image of each cow to automatically detect multi-object cattle rumination. To verify the feasibility of this method, the algorithm was tested on multi-object cattle rumination videos, and the results were compared with the results produced by human observation. The experimental results showed that the average error in rumination time was 5.902% and the average error in the number of chews was 8.126%. The rumination identification and calculation of rumination information only need to be performed by computers automatically with no manual intervention. It could provide a new contactless rumination identification method for multi-cattle, which provided technical support for smart pasture.

Acoustic communication and imaging sonar guided AUV docking: system infrastructure, docking methodology and lake trials

Autonomous underwater vehicle (AUV) docking technique attracts attention which leverages the long-term, on-station vehicle launch and recovery for the underwater missions. In this paper, an acoustic communication and imaging sonar guided docking method is proposed to improve the docking accuracy for AUV. In docking system, multibeam forward looking sonar (MFLS) is deployed on the submerged docking station to reduce the noise in acoustic images rather than being fixed on AUV, and the ultra short base line (USBL) transceiver with acoustic communication modules is used to achieve positioning while AUV is out of the view of MFLS. For the docking methodology, a dedicated three-stage docking strategy is presented to guide AUV to the docking station. In the homing stage, AUV navigates to a preset homing place not very far from the docking station via the way-point guidance and self-positioning of the on-board inertial navigation system (INS). In the next stage, USBL provides the relative position for AUV controller via acoustic channel. Moreover, an exponential decay model based filtering algorithm is proposed to eliminate the positioning outliers of USBL. It is expected that AUV will be guided into the perceptual field of MFLS in this stage. In the final stage, the docking accuracy in short range should be improved while USBL may generate outliers due to the louder noise. In this case, the relative positioning is implemented by using kernelized correlation filter (KCF), which is a fast objective tracking algorithm to keep track of the AUV feature in acoustic image until AUV navigates into the docking station or out of the view of sensor. To validate feasibility and performance of the proposed system infrastructure and docking methodology, the docking results are analyzed by contrast with the experiments only use USBL during lake trials.

A model adaptive updating kernel correlation filter tracker with deep CNN features

The tracker based on correlation filter shows excellent performance in tracking accuracy and running speed. However, the models of correlated filter trackers are always updated with fixed weights, which can degrade the tracking performance when the target in a variety of challenging scenarios. In this paper, we present a model adaptive updating method based on a fuzzy system, which can set different updating weights on each frame to effectively deal with the challenging scenarios in the tracking process. Attractively, this method can be perfectly used in all trackers based on correlation filtering. In addition, we combine deep CNN features that can describe target semantics with HOG features that have spatial descriptions. Using their complementarity to target descriptions, we establish HOG-based filter model and CNN-based filter model. To two response maps of the models, we propose a different fusion strategy based on quality measurement of tracking results, which can balance the accuracy and robustness of the tracker. Experiments on OTB-2013, OTB-2015, benchmark videos, and VOT2018 dataset show that our tracker (called MACF) is effective and exhibits competitive results compared with the recent state-of-the-art (SOTA) trackers.