Visual Object Tracking Method Research Articles

Transformer object tracking algorithm based on spatiotemporal template focal attention

Existing visual object tracking methods only use the target area of the first frame as a template, which makes it easy to fail in fast-changing and complex backgrounds. To solve this problem, this paper proposes a Transformer-based object tracking algorithm that focuses on the target focus information in the template and dynamically updates the template features. In order to reduce the interference of background information on attention, this algorithm uses a sparse Transformer module to achieve the interaction of feature information; a template focus attention module is also proposed to focus on the dynamic template features and the initial template features to retain the highly reliable feature information in the initial template. Experimental results show that the success rate and accuracy of this algorithm in the 0TB100 benchmark test are and respectively 70.9%, 91.6%and the success rate and accuracy are improved by 4.11%and respectively compared with the similar template update algorithm STARK 3.27%. This algorithm effectively addresses the limitations of existing visual object tracking methods and improves the accuracy and robustness of tracking.

Spatio-temporal interactive fusion based visual object tracking method

Visual object tracking tasks often struggle with utilizing inter-frame correlation information and handling challenges like local occlusion, deformations, and background interference. To address these issues, this paper proposes a spatio-temporal interactive fusion (STIF) based visual object tracking method. The goal is to fully utilize spatio-temporal background information, enhance feature representation for object recognition, improve tracking accuracy, adapt to object changes, and reduce model drift. The proposed method incorporates feature-enhanced networks in both temporal and spatial dimensions. It leverages spatio-temporal background information to extract salient features that contribute to improved object recognition and tracking accuracy. Additionally, the model’s adaptability to object changes is enhanced, and model drift is minimized. A spatio-temporal interactive fusion network is employed to learn a similarity metric between the memory frame and the query frame by utilizing feature enhancement. This fusion network effectively filters out stronger feature representations through the interactive fusion of information. The proposed tracking method is evaluated on four challenging public datasets. The results demonstrate that the method achieves state-of-the-art (SOTA) performance and significantly improves tracking accuracy in complex scenarios affected by local occlusion, deformations, and background interference. Finally, the method achieves a remarkable success rate of 78.8% on TrackingNet, a large-scale tracking dataset.

Cascaded Denoising Transformer for UAV Nighttime Tracking

The automation of unmanned aerial vehicles (UAVs) has been greatly promoted by visual object tracking methods with onboard cameras. However, the random and complicated real noise produced by the cameras seriously hinders the performance of state-of-the-art (SOTA) UAV trackers, especially in low-illumination environments. To address this issue, this work proposes an efficient plug-and-play cascaded denoising Transformer (CDT) to suppress cluttered and complex real noise, thereby boosting UAV tracking performance. Specifically, the novel U -shaped cascaded denoising network is designed with a streamlined structure for efficient computation. Additionally, shallow feature deepening (SFD) encoder and multi-feature collaboration (MFC) decoder are constructed based on multi-head transposed self-attention (MTSA) and multi-head transposed cross-attention (MTCA), respectively. A nested residual feed-forward network (NRFN) is developed to focus more on high-frequency information represented by noise. Extensive evaluation and test experiments demonstrate that the proposed CDT has a remarkable denoising effect and improves UAV nighttime tracking performance.

Adaptive chaotic sampling particle filter to handle occlusion and fast motion in visual object tracking

We present a new particle filter method for visual object tracking to effectively handle occlusion and fast motion. The proposed approach uses a chaotic local search to model irregular motion and, compared to ordinary particle filter approaches, requires a lower number of particles. Furthermore, a new chaotic sampling procedure is used to force particles to specific areas with high values of the likelihood function with maximum diversity and a histogram of dynamical information is introduced to represent the motion over successive frames based on state space reconstruction. Finally, a new criterion is proposed to distinguish occlusion and out-of-view for appearance updating. We present numerical experiments demonstrating that the developed framework outperforms other state-of-the-art approaches dealing with irregular motions and uncertainties. According to the results on BOBOT, OTB100, OTB2013, and VOT2018, compared with traditional approaches, including methods based on deep and reinforcement learning, correlation filters, and Siamese neural networks, the proposed strategy leads to much closer convergence to the true target state, increasing the tracking accuracy. Finally, we prove analytically the convergence of the proposed method.

A Siamese Network for real-time object tracking on CPU

<h2>Abstract</h2> Visual object tracking methods depend upon deep networks that can hardly meet real-time processing requirements on mobile platforms with limited computing resources. In this work, we propose a real-time object tracking framework by enhancing a lightweight feature pyramid network with Transformer architecture to construct a robust target-specific appearance model efficiently. We further introduce the pooling attention module to avoid the computation and memory intensity while fusing pyramid features with the Transformer. The optimized tracker operates over 45 Hz on a single CPU, allowing researchers to deploy it on any mobile device with limited power resources.

Biogeography based optimization method for robust visual object tracking

Moving object tracking is one of the applied fields in artificial intelligence and robotic. The main objective of object tracking is to detect and locate targets in video frames of real scenes. Although various methods have been proposed for object tracking so far, tracking in challenging conditions remains an open issue. Recently, different evolutionary and heuristics algorithms like swarm intelligence have been used to address the tracking challenges, which have shown promising performance. In this paper, a new approach based on modified biogeography based optimization (mBBO) method is introduced. The BBO algorithm includes migration and mutation steps. In the migration phase, the search space is properly explored by sharing information between habitats and weaker solutions to improve their position. On the other hand, the mutation phase leads to diversity and change in solutions. In this algorithm, the elitist method has been also used to keep better solutions. The performance of modified BBO tracker has been evaluated on benchmark video datasets and compared with several other tracking methods. Experimental results demonstrate that the proposed method estimates the location of targets with high accuracy and achieves better performance and robustness compared to other trackers.

Learning Enhanced Feature Responses for Visual Object Tracking.

Visual object tracking is an important topic in computer vision, which has successfully utilized pretrained convolutional neural networks, such as VGG and ResNet. However, the features extracted by these pretrained models are high dimensional, and the redundant feature channels reduce target localization and scale estimation precision, leading to tracking drifting. In this paper, a novel visual object tracking method, called learning enhanced feature responses tracking (LEFRT), is proposed, which adopts the target-specific features to enhance target localization and scale estimation responses. First, a channel attention module, called target-specific network (TSNet), is presented to reduce the redundant feature channels. Secondly, the scale estimation network (SCENet) is introduced to extract spatial structural features to generate a more precise response for the scale estimation. Extensive experiments on six tracking benchmarks, including LaSOT, GOT-10k, TrackingNet, OTB-2013, OTB-2015, and TC-128, demonstrate that the proposed algorithm can effectively improve the precision and speed of visual object tracking. LEFRT achieves 90.4% precision and a 71.2% success rate on the OTB-2015 dataset, improving the tracking methods based on the pretrained features.

Co-saliency-regularized correlation filter for object tracking

Correlation filtering (CF) has emerged as one of the best object-tracking frameworks, achieving a good balance between tracking accuracy and speed. Based on this framework, many successful trackers have been developed. We propose a new co-saliency regularization method for visual object tracking based on the correlation filter, called the CRCF. To the best our knowledge, this is the first application of co-saliency regularization to CF-based tracking, in the context of widely used co-saliency detection and regularization methods. In the CRCF, the co-saliency information is extracted and introduced into the regularization component in the relevant tracking framework. The model first calculates the salient regions based on the feature differences between salient objects and background information. A co-saliency map is generated for the global correspondence of implicit learning between multiple images. To avoid traversing all the pixels, we use SLIC superpixel clustering to extract the objects saliency information, thus significantly reducing the computational overhead in the image processing step while retaining the image features and structural information. Finally, during the tracking process, it is possible to dynamically learn the co-saliency map configuration, highlight relevant object areas, and mitigate the impact of the lack of discriminatory representations on the performance. Quantitative and qualitative experimental results on OTB-2015, UAV123, LaSOT, GOT-10k and VOT-2018 datasets show that our CRCF outperforms the latest trackers.

Selective part‐based correlation filter tracking algorithm with reinforcement learning

In visual object tracking methods, improving both the run time and the accuracy in the face of complex situations has always been an important issue. Many complex tracking algorithms, such as part-based algorithms, have better accuracy when facing occlusions, but they have much greater computational complexity. In response to the above problems, this paper proposes a selective part-based correlation filter (SPCF) tracking algorithm with a reinforcement learning to achieve more stable and efficient tracking of targets. First, according to the conditions of the response map of the correlation filter (CF), the entire tracking process is divided into three states: simple environments, complex environments, and harsh environments. Second, this paper uses reinforcement learning to determine the states of frames in different situations to improve the tracking effect of the algorithm. Third, the process of the online selection of states is transformed into a Markov decision process (MDP), where the policy learning of the MDP is achieved by reinforcement learning. Additionally, different strategies are used to track a target in different states: the overall filter is used to increase the speed in simple environments; part-based filters are used to improve the accuracy in complex environments; and in harsh environments where the target completely disappears, a redetection algorithm is used to find the target when it reappears. Finally, the performance of the tracking algorithm is verified on the VOT2018, OTB-2015, and LaSOT datasets.

Multitarget Tracking Using Siamese Neural Networks

In this article, we detect and track visual objects by using Siamese network or twin neural network. The Siamese network is constructed to classify moving objects based on the associations of object detection network and object tracking network, which are thought of as the two branches of the twin neural network. The proposed tracking method was designed for single-target tracking, which implements multitarget tracking by using deep neural networks and object detection. The contributions of this article are stated as follows. First, we implement the proposed method for visual object tracking based on multiclass classification using deep neural networks. Then, we attain multitarget tracking by combining the object detection network and the single-target tracking network. Next, we uplift the tracking performance by fusing the outcomes of the object detection network and object tracking network. Finally, we speculate on the object occlusion problem based on IoU and similarity score, which effectively diminish the influence of this issue in multitarget tracking.

Aberrance suppressed spatio-temporal correlation filters for visual object tracking

The objective of the present study is to design a correlation filter-based tracking method for robust visual object tracking. In the literature, numerous tracking methods have been proposed based on discriminative correlation filter (DCF) and obtained impressive performance. However, existing algorithms still face difficulties such as partial occlusion, clutter background, uncertainties, boundary effects (especially when the target search area is small) and other challenging visual factors. Furthermore, during the target detection process, the sudden changes in objects caused by illumination variations and partial/full occlusion degrade the performance. To tackle the drawbacks mentioned earlier, we propose a tracking algorithm concerning the aberrance suppressed correlation filters with spatio-temporal information for visual tracking. Specifically, we introduce a spatial regularization term into the correlation filter to suppresses the boundary effects. Following that, a temporal regularization is adopted into the DCF-based framework to achieve a more robust appearance model and further enhance the tracking performance. In addition, we introduce an approach to suppress the aberrance in response maps caused by the sudden changes. Technically, our proposed method can be directly solved by using the alternating direction method of multipliers (ADMM) technique with a low computational cost. Finally, extensive experimental results on OTB2013, OTB2015, TempleColor128 and UAV123 datasets demonstrate that the proposed method performs favorably against state-of-the-art methods.

JROTM: Jointly reinforced object tracking with temporal content reference and motion guidance

Although the existing visual object tracking methods have made noticeable progress, many chokepoints remain unsettled. One of the most significant performance bottlenecks is the drastic change of the target, such as fast motion, scale variation, appearance changes, etc. Focusing on this major barrier, we propose our object tracking method, JROTM. It adopts the two-stage tracking-by-detection strategy, and it is reinforced by the temporal content reference (TCR) module and the motion guidance (MG) module. TCR precisely provides auxiliary frame-wise information of two consecutive frames of a video. With this information, the rapid change of the object’s appearance and position is retarded. MG module accurately offers more target position hints at the current frame. Our proposed JROTM is evaluated in multiple popular tracking benchmark, including OTB-100, TempleColor-128, and VOT-2016. It outperforms the state-of-the-art tracking methods by a non-neglectable margin.

Siamese networks with distractor-reduction method for long-term visual object tracking

Many trackers which divide the tracking process into two stages have recently been proposed to solve the problem of long-term tracking. Their outstanding performance makes them become one of the mainstream algorithms of long-term tracking. To further improve the performance of two-stage tracking algorithms, some improvements are proposed in this paper. (a) A hard negative mining method is proposed. It can optimize the training process of the verification network and bridge the gap between the two sub-networks. (b) The architecture of the verification network is designed as a Siamese structure; therefore, the semantic ambiguity in classification can be alleviated. Extensive experiments performed on benchmarks demonstrate that the proposed approach significantly outperforms the state-of-the-art methods, yielding 7% relative gain in the VOT2018-LT dataset and 14.2% relative gain in the OxUvA dataset.

CNN-based single object detection and tracking in videos and its application to drone detection

This paper presents convolutional neural network (CNN)-based single object detection and tracking algorithms. CNN-based object detection methods are directly applicable to static images, but not to videos. On the other hand, model-free visual object tracking methods cannot detect an object until a ground truth bounding box of the target is provided. Moreover, many annotated video datasets of the target object are required to train both the object detectors and visual trackers. In this work, three simple yet effective object detection and tracking algorithms for videos are proposed to efficiently combine a state-of-the-art object detector and visual tracker for circumstances in which only a few static images of the target are available for training. The proposed algorithms are tested using a drone detection task and the experimental results demonstrated their effectiveness.

SiamMN: Siamese modulation network for visual object tracking

Visual object tracking methods based on Siamese network are often difficult to distinguish objects with the same semantic or similar appearance as tracking target in tracking process due to the lack of discriminating strategies for the confusing objects. We propose a visual object tracking method based on Siamese modulation network. It takes the given bounding box in the target frame and the current frame as input, and fuses these multi-layer convolutional features to obtain more target appearance information of bounding box and the current frame. The feature modulator generates feature modulation vector based on the given bounding box to enhance visual appearance information of target instance in multi-layer feature of the current frame, so as to make target instance obtain higher score in response map of region proposal network, and thus realize target instance-specific tracking task. Experiments on two public benchmark datasets, OTB2015 and VOT2018, show that the proposed tracker has a competitive performance among other state-of-the art trackers.

Robust visual tracking via part-based model

In this paper, we propose a novel visual object tracking method using a part based appearance model. First, a local kernel feature is developed to encode edge information of patches. Next, bounding box of the target is divided into multiple parts. Then, each part uses correlation filter based tracking to predict position in the next frame. The matrix cosine similarity is utilized to measure reliabilities of the patches. Finally, optimal target location is predicted via maximizing the likelihood, which is obtained by adaptively fusing the reliable patches locations. Experimental results illustrate that our algorithm outperforms state-of-the-art tracking methods significantly in terms of accuracy and robustness.

Gaussian-response correlation filter for robust visual object tracking

This paper presents a novel correlation filter-based tracking method for robust visual object tracking in the presence of partial occlusion, large-scale variation and model drift. To do this, first, we develop a correlation filter for predicting the target location based on the distribution of correlation response. In this formulation, the correlation response of the target image follows Gaussian distribution to estimate the target location efficiently. Second, the constraints are derived using kernel ridge regression to mitigate the target failure in object tracking. Third, we propose an adaptive scale estimation method to detect the target scale changes during the tracking. In addition, two feature integration is elaborately designed to improve the discriminative strength of the correlation filter. Finally, extensive experimental results on OTB2013, OTB2015, TempleColor128 and UAV123 datasets demonstrate that the proposed method performs favourably against several state-of-the-art methods.

Discriminative object tracking with subspace representation

Visual object tracking is a core research area in the field of pattern recognition and computer vision. It becomes one of the most significant tasks in computer vision application. But tracking of a visual object is not an easy task as it is always restricted by appearance change, illumination, occlusion and so on. Object tracking based on principal component analysis (PCA) is one of the most effective tracking methods as it can handle the different challenging problems of the tracking algorithm. But in this PCA-based tracking method, the background pixels are also included in the subspace representation of the target object, and so this method cannot overcome all the problems of tracking. In this work, a robust visual object tracking method is proposed by introducing sparse discriminative classifier (SDC) feature selection in PCA subspace representation. The SDC method is utilized to extract the target object from the template image target by removing the background pixels which is unnecessary for tracking task without much computational complexity. The PCA algorithm adequately represents a presentation model of the target object and account of occlusion with trivial template. Qualitative and quantitative analysis of different diverse videos shows that the newly proposed method outperforms the other existing state-of-the-art tracking algorithm.

A Learning Frequency-Aware Feature Siamese Network for Real-Time Visual Tracking

Visual object tracking by Siamese networks has achieved favorable performance in accuracy and speed. However, the features used in Siamese networks have spatially redundant information, which increases computation and limits the discriminative ability of Siamese networks. Addressing this issue, we present a novel frequency-aware feature (FAF) method for robust visual object tracking in complex scenes. Unlike previous works, which select features from different channels or layers, the proposed method factorizes the feature map into multi-frequency and reduces the low-frequency information that is spatially redundant. By reducing the low-frequency map’s resolution, the computation is saved and the receptive field of the layer is also increased to obtain more discriminative information. To further improve the performance of the FAF, we design an innovative data-independent augmentation for object tracking to improve the discriminative ability of tracker, which enhanced linear representation among training samples by convex combinations of the images and tags. Finally, a joint judgment strategy is proposed to adjust the bounding box result that combines intersection-over-union (IoU) and classification scores to improve tracking accuracy. Extensive experiments on 5 challenging benchmarks demonstrate that our FAF method performs favorably against SOTA tracking methods while running around 45 frames per second.

Visual object tracking using sparse context-aware spatio-temporal correlation filter

This paper presents a novel sparse context-aware spatio-temporal correlation filter tracker (SCAST) method for robust visual object tracking. Different from the existing trackers, this paper introduce an l1 multi-scale regularization parameter-based correlation filter that reduces the boundary effect due to partial occlusions, illumination and scale variations. At each iteration, the l1 regularization parameter is updated through spatial knowledge of each correlation filter coefficient. Besides, the contextual information acquired from the target region can lead to determining the accurate localization of the target. Moreover, contextual information has combined with spatio-temporal factor to achieve the better performance. Further, an objective function is designed with system constraints to ensure the applicability of the model and the optimal solution is derived by utilizing the alternating direction method of multiplier, which leads to low computational cost. Finally, the feasibility and superiority of proposed tracker algorithm is evaluated through three benchmark dataset: OTB-2013, OTB-2015, and TempleColor-128.