scholarly journals Exploiting Spatial Invariance for Scalable Unsupervised Object Tracking

2020 ◽  
Vol 34 (04) ◽  
pp. 3684-3692
Author(s):  
Eric Crawford ◽  
Joelle Pineau
Keyword(s):  
Object Tracking ◽  
Spatial Attention ◽  
Intelligent Agent ◽  
Current Work ◽  
Visual World ◽  
Reasoning Process ◽  
Series Of Experiments ◽  
Spatial Invariance ◽  
Tracking Objects ◽  
Object Level

The ability to detect and track objects in the visual world is a crucial skill for any intelligent agent, as it is a necessary precursor to any object-level reasoning process. Moreover, it is important that agents learn to track objects without supervision (i.e. without access to annotated training videos) since this will allow agents to begin operating in new environments with minimal human assistance. The task of learning to discover and track objects in videos, which we call unsupervised object tracking, has grown in prominence in recent years; however, most architectures that address it still struggle to deal with large scenes containing many objects. In the current work, we propose an architecture that scales well to the large-scene, many-object setting by employing spatially invariant computations (convolutions and spatial attention) and representations (a spatially local object specification scheme). In a series of experiments, we demonstrate a number of attractive features of our architecture; most notably, that it outperforms competing methods at tracking objects in cluttered scenes with many objects, and that it can generalize well to videos that are larger and/or contain more objects than videos encountered during training.

On Spatial Attention in Object Tracking using a Synthetic Aperture Camera Array

2012 ◽  
Author(s):  
Christoph Walter ◽  
Maik Poggendorf ◽  
Norbert Elkmann ◽  
Felix Penzlin
Keyword(s):  
Object Tracking ◽  
Spatial Attention ◽  
Synthetic Aperture ◽  
Camera Array

scholarly journals Learning Soft Mask Based Feature Fusion with Channel and Spatial Attention for Robust Visual Object Tracking

Sensors ◽  
2020 ◽  
Vol 20 (14) ◽  
pp. 4021 ◽  
Author(s):  
Mustansar Fiaz ◽  
Arif Mahmood ◽  
Soon Ki Jung
Keyword(s):  
Object Tracking ◽  
Spatial Attention ◽  
Feature Fusion ◽  
State Of The Art ◽  
Feature Representation ◽  
Visual Object ◽  
Target Feature ◽  
Visual Object Tracking ◽  
Low Level ◽  
Benchmark Datasets

We propose to improve the visual object tracking by introducing a soft mask based low-level feature fusion technique. The proposed technique is further strengthened by integrating channel and spatial attention mechanisms. The proposed approach is integrated within a Siamese framework to demonstrate its effectiveness for visual object tracking. The proposed soft mask is used to give more importance to the target regions as compared to the other regions to enable effective target feature representation and to increase discriminative power. The low-level feature fusion improves the tracker robustness against distractors. The channel attention is used to identify more discriminative channels for better target representation. The spatial attention complements the soft mask based approach to better localize the target objects in challenging tracking scenarios. We evaluated our proposed approach over five publicly available benchmark datasets and performed extensive comparisons with 39 state-of-the-art tracking algorithms. The proposed tracker demonstrates excellent performance compared to the existing state-of-the-art trackers.

scholarly journals Multibandwidth Kernel-Based Object Tracking

2010 ◽  
Vol 2010 ◽  
pp. 1-15 ◽  
Author(s):  
Aras Dargazany ◽  
Ali Soleimani ◽  
Alireza Ahmadyfard
Keyword(s):  
Object Tracking ◽  
Mean Shift ◽  
Initial Point ◽  
Sampling Procedure ◽  
Target Movement ◽  
Global Mode ◽  
Local Maxima ◽  
Multiple Kernel ◽  
Global Modes ◽  
Tracking Objects

Object tracking using Mean Shift (MS) has been attracting considerable attention recently. In this paper, we try to deal with one of its shortcoming. Mean shift is designed to find local maxima for tracking objects. Therefore, in large target movement between two consecutive frames, the local and global modes are not the same as previous frames so that Mean Shift tracker may fail in tracking the desired object via localizing the global mode. To overcome this problem, a multibandwidth procedure is proposed to help conventional MS tracker reach the global mode of the density function using any staring points. This gradually smoothening procedure is called Multi Bandwidth Mean Shift (MBMS) which in fact smoothens the Kernel Function through a multiple kernel-based sampling procedure automatically. Since it is important for us to have less computational complexity for real-time applications, we try to decrease the number of iterations to reach the global mode. Based on our results, this proposed version of MS enables us to track an object with the same initial point much faster than conventional MS tracker.

scholarly journals Enhanced Multiple-Object Tracking Using Delay Processing and Binary-Channel Verification

2019 ◽  
Vol 9 (22) ◽  
pp. 4771 ◽  
Author(s):  
Muyu Li ◽  
Xin He ◽  
Zhonghui Wei ◽  
Jun Wang ◽  
Zhiya Mu ◽  
...  
Keyword(s):  
Object Tracking ◽  
Multiple Object Tracking ◽  
Single Object ◽  
Complex Scene ◽  
Whole Process ◽  
Multiple Object ◽  
Binary Channel ◽  
Complex Scenes ◽  
Verification Model ◽  
Tracking Objects

Tracking objects over time, i.e., identity (ID) consistency, is important when dealing with multiple object tracking (MOT). Especially in complex scenes with occlusion and interaction of objects this is challenging. Significant improvements in single object tracking (SOT) methods have inspired the introduction of SOT to MOT to improve the robustness, that is, maintaining object identities as long as possible, as well as helping alleviate the limitations from imperfect detections. SOT methods are constantly generalized to capture appearance changes of the object, and designed to efficiently distinguish the object from the background. Hence, simply extending SOT to a MOT scenario, which consists of a complex scene with spatially mixed, occluded, and similar objects, will encounter problems in computational efficiency and drifted results. To address this issue, we propose a binary-channel verification model that deeply excavates the potential of SOT in refining the representation while maintaining the identities of the object. In particular, we construct an integrated model that jointly processes the previous information of existing objects and new incoming detections, by using a unified correlation filter through the whole process to maintain consistency. A delay processing strategy consisting of the three parts—attaching, re-initialization, and re-claiming—is proposed to tackle drifted results caused by occlusion. Avoiding the fuzzy appearance features of complex scenes in MOT, this strategy can improve the ability to distinguish specific objects from each other without contaminating the fragile training space of a single object tracker, which is the main cause of the drift results. We demonstrate the effectiveness of our proposed approach on the MOT17 challenge benchmarks. Our approach shows better overall ID consistency performance in comparison with previous works.

Multiple Object Detection and Tracking in Complex Background

2017 ◽  
Vol 31 (02) ◽  
pp. 1755003 ◽  
Author(s):  
Yingdong Ma ◽  
Yuchen Liu ◽  
Shuai Liu ◽  
Zhibin Zhang
Keyword(s):  
Object Tracking ◽  
Model Updating ◽  
Expectation Maximization Algorithm ◽  
Gaussian Mixture ◽  
Multiple Targets ◽  
Detection And Tracking ◽  
Multiple Object ◽  
Complex Background ◽  
Tracking Objects ◽  
Multiple Particle

Multiple object tracking is a fundamental step for many computer vision applications. However, detecting and tracking objects in complex background is still a challenging task. This paper proposes an approach, which combines an improved Gaussian mixture modeling (GMM) with multiple particle filters (MPFs) for automatic multiple targets detecting and tracking. For GMM, we make improvement on GMM in the phase of model updating by using the expectation maximization algorithm and [Formula: see text] recent frames with weight parameters of Gaussian distributions. In the tracking stage, we integrate multiple features of targets, including color, edge and depth, into MPFs to improve the performance of object tracking. By comparing with various particle filter approaches, the experimental results show that our approach can track multiple targets in complex backgrounds automatically and accurately.

scholarly journals Comparative Analysis of Tracking Objects Using Optical Flow and Background Estimation on Silent Camera

2018 ◽  
pp. 153-162
Author(s):  
Wahyu Supriyatin ◽  
Winda Widya Ariestya ◽  
Ida Astuti
Keyword(s):  
Computer Vision ◽  
Optical Flow ◽  
Object Tracking ◽  
Estimation Method ◽  
Flow Estimation ◽  
Background Estimation ◽  
Existing Problems ◽  
Parameter Values ◽  
Tracking Objects ◽  
Better Than

Tracking and object is one of the utilizations on the field of the computer vision application. Object tracking utilization as a computer vision in this study is used to identify objects which exist within a frame and calculate the number of objects passing within a frame. The utilization of computer vision in various fields of application can be used to solve the existing problems. The method used in object tracking is by comparison between optical flow estimation method with background method. The test is conducted by using a still camera for both methods by making changes to the parameter values used as a reference. The results of the tests, conducted on the three video objects by comparing the two methods show a Total Recorded Time better than those of the background estimation method, being smaller than 100 seconds. Testing both methods successfully identifies the object tracking and calculates the number of passing cars.

scholarly journals Flying object tracking and classification of military versus nonmilitary aircraft

2020 ◽  
Vol 9 (4) ◽  
pp. 1394-1403
Author(s):  
Ehsan Akbari Sekehravani ◽  
Eduard Babulak ◽  
Mehdi Masoodi
Keyword(s):  
Machine Vision ◽  
Particle Filter ◽  
Object Tracking ◽  
Moving Objects ◽  
Center Of Mass ◽  
Speed Of Processing ◽  
Detection And Tracking ◽  
Particle Filter Algorithm ◽  
Tracking Objects

Tracking of moving objects in a sequence of images is one of the important and functional branches of machine vision technology. Detection and tracking of a flying object with unknown features are important issues in detecting and tracking objects. This paper consists of two basic parts. The first part involves tracking multiple flying objects. At first, flying objects are detected and tracked, using the particle filter algorithm. The second part is to classify tracked objects (military or nonmilitary), based on four criteria; Size (center of mass) of objects, object speed vector, the direction of motion of objects, and thermal imagery identifies the type of tracked flying objects. To demonstrate the efficiency and the strength of the algorithm and the above system, several scenarios in different videos have been investigated that include challenges such as the number of objects (aircraft), different paths, the diverse directions of motion, different speeds and various objects. One of the most important challenges is the speed of processing and the angle of imaging.

Sign in / Sign up

Export Citation Format

Share Document