In this paper, we propose and study a novel visual object tracking approach based on convolutional networks and recurrent networks. The proposed approach is distinct from the existing approaches to visual object tracking, such as filtering-based ones and tracking-by-detection ones, in the sense that the tracking system is explicitly trained off-line to track anonymous objects in a noisy environment. The proposed visual tracking model is end-to-end trainable, minimizing any adversarial effect from mismatches in object representation and between the true underlying dynamics and learning dynamics. We empirically show that the proposed tracking approach works well in various scenarios by generating artificial video sequences with varying conditions; the number of objects, amount of noise and the match between the training shapes and test shapes.
Multi-object tracking remains challenging due to frequent occurrence of occlusions and outliers. In order to handle this problem, we propose an Approximation-Shrink Scheme for sequential optimization. This scheme is realized by introducing an Ambiguity-Clearness Graph to avoid conflicts and maintain sequence independent, as well as a sliding window optimization framework to constrain the size of state space and guarantee convergence. Based on this window-wise framework, the states of targets are clustered in a self-organizing manner. Moreover, we show that the traditional online and batch tracking methods can be embraced by the window-wise framework. Experiments indicate that with only a small window, the optimization performance can be much better than online methods and approach to batch methods.
Given a set of detections, detected at each time instant independently, we investigate how to associate them across time. This is done by propagating labels on a set of graphs, each graph capturing how either the spatio-temporal or the appearance cues promote the assignment of identical or distinct labels to a pair of detections. The graph construction is motivated by a locally linear embedding of the detection features. Interestingly, the neighborhood of a node in appearance graph is defined to include all the nodes for which the appearance feature is available (even if they are temporally distant). This gives our framework the uncommon ability to exploit the appearance features that are available only sporadically. Once the graphs have been defined, multi-object tracking is formulated as the problem of finding a label assignment that is consistent with the constraints captured each graph, which results into a difference of convex (DC) program. We propose to decompose the global objective function into node-wise sub-problems. This not only allows a computationally efficient solution, but also supports an incremental and scalable construction of the graph, thereby making the framework applicable to large graphs and practical tracking scenarios. Moreover, it opens the possibility of parallel implementation.
Deep learning has recently achieved very promising results in a wide range of areas such as computer vision, speech recognition and natural language processing. It aims to learn hierarchical representations of data by using deep architecture models. In a smart city, a lot of data (e.g. videos captured from many distributed sensors) need to be automatically processed and analyzed. In this paper, we review the deep learning algorithms applied to video analytics of smart city in terms of different research topics: object detection, object tracking, face recognition, image classification and scene labeling.
We propose a framework called ReFInE to directly obtain integral image estimates from a very small number of spatially multiplexed measurements of the scene without iterative reconstruction of any auxiliary image, and demonstrate their practical utility in visual object tracking. Specifically, we design measurement matrices which are tailored to facilitate extremely fast estimation of the integral image, by using a single-shot linear operation on the measured vector. Leveraging a prior model for the images, we formulate a nuclear norm minimization problem with second order conic constraints to jointly obtain the measurement matrix and the linear operator. Through qualitative and quantitative experiments, we show that high quality integral image estimates can be obtained using our framework at very low measurement rates. Further, on a standard dataset of 50 videos, we present object tracking results which are comparable to the state-of-the-art methods, even at an extremely low measurement rate of 1%.
Humans use context and scene knowledge to easily localize moving objects in conditions of complex illumination changes, scene clutter and occlusions. In this paper, we present a method to leverage human knowledge in the form of annotated video libraries in a novel search and retrieval based setting to track objects in unseen video sequences. For every video sequence, a document that represents motion information is generated. Documents of the unseen video are queried against the library at multiple scales to find videos with similar motion characteristics. This provides us with coarse localization of objects in the unseen video. We further adapt these retrieved object locations to the new video using an efficient warping scheme. The proposed method is validated on in-the-wild video surveillance datasets where we outperform state-of-the-art appearance-based trackers. We also introduce a new challenging dataset with complex object appearance changes.
We propose an adaptive tracking algorithm where the object is modelled as a continuously updated bag of affine subspaces, with each subspace constructed from the object's appearance over several consecutive frames. In contrast to linear subspaces, affine subspaces explicitly model the origin of subspaces. Furthermore, instead of using a brittle point-to-subspace distance during the search for the object in a new frame, we propose to use a subspace-to-subspace distance by representing candidate image areas also as affine subspaces. Distances between subspaces are then obtained by exploiting the non-Euclidean geometry of Grassmann manifolds. Experiments on challenging videos (containing object occlusions, deformations, as well as variations in pose and illumination) indicate that the proposed method achieves higher tracking accuracy than several recent discriminative trackers.
We consider the problem of model-based 3D-tracking of objects given dense depth images as input. Two difficulties preclude the application of a standard Gaussian filter to this problem. First of all, depth sensors are characterized by fat-tailed measurement noise. To address this issue, we show how a recently published robustification method for Gaussian filters can be applied to the problem at hand. Thereby, we avoid using heuristic outlier detection methods that simply reject measurements if they do not match the model. Secondly, the computational cost of the standard Gaussian filter is prohibitive due to the high-dimensional measurement, i.e. the depth image. To address this problem, we propose an approximation to reduce the computational complexity of the filter. In quantitative experiments on real data we show how our method clearly outperforms the standard Gaussian filter. Furthermore, we compare its performance to a particle-filter-based tracking method, and observe comparable computational efficiency and improved accuracy and smoothness of the estimates.
In this paper, we propose a novel method for visual object tracking called HMMTxD. The method fuses observations from complementary out-of-the box trackers and a detector by utilizing a hidden Markov model whose latent states correspond to a binary vector expressing the failure of individual trackers. The Markov model is trained in an unsupervised way, relying on an online learned detector to provide a source of tracker-independent information for a modified Baum- Welch algorithm that updates the model w.r.t. the partially annotated data. We show the effectiveness of the proposed method on combination of two and three tracking algorithms. The performance of HMMTxD is evaluated on two standard benchmarks (CVPR2013 and VOT) and on a rich collection of 77 publicly available sequences. The HMMTxD outperforms the state-of-the-art, often significantly, on all datasets in almost all criteria.
This paper presents to the best of our knowledge the first end-to-end object tracking approach which directly maps from raw sensor input to object tracks in sensor space without requiring any feature engineering or system identification in the form of plant or sensor models. Specifically, our system accepts a stream of raw sensor data at one end and, in real-time, produces an estimate of the entire environment state at the output including even occluded objects. We achieve this by framing the problem as a deep learning task and exploit sequence models in the form of recurrent neural networks to learn a mapping from sensor measurements to object tracks. In particular, we propose a learning method based on a form of input dropout which allows learning in an unsupervised manner, only based on raw, occluded sensor data without access to ground-truth annotations. We demonstrate our approach using a synthetic dataset designed to mimic the task of tracking objects in 2D laser data -- as commonly encountered in robotics applications -- and show that it learns to track many dynamic objects despite occlusions and the presence of sensor noise.
The problem of visual tracking evaluation is sporting a large variety of performance measures, and largely suffers from lack of consensus about which measures should be used in experiments. This makes the cross-paper tracker comparison difficult. Furthermore, as some measures may be less effective than others, the tracking results may be skewed or biased towards particular tracking aspects. In this paper we revisit the popular performance measures and tracker performance visualizations and analyze them theoretically and experimentally. We show that several measures are equivalent from the point of information they provide for tracker comparison and, crucially, that some are more brittle than the others. Based on our analysis we narrow down the set of potential measures to only two complementary ones, describing accuracy and robustness, thus pushing towards homogenization of the tracker evaluation methodology. These two measures can be intuitively interpreted and visualized and have been employed by the recent Visual Object Tracking (VOT) challenges as the foundation for the evaluation methodology.
We present an on-line 3D visual object tracking framework for monocular cameras by incorporating spatial knowledge and uncertainty from semantic mapping along with high frequency measurements from visual odometry. Using a combination of vision and odometry that are tightly integrated we can increase the overall performance of object based tracking for semantic mapping. We present a framework for integration of the two data-sources into a coherent framework through information based fusion/arbitration. We demonstrate the framework in the context of OmniMapper[1] and present results on 6 challenging sequences over multiple objects compared to data obtained from a motion capture systems. We are able to achieve a mean error of 0.23m for per frame tracking showing 9% relative error less than state of the art tracker.
Global motion compensation (GMC) removes the impact of camera motion and creates a video in which the background appears static over the progression of time. Various vision problems, such as human activity recognition, background reconstruction, and multi-object tracking can benefit from GMC. Existing GMC algorithms rely on sequentially processing consecutive frames, by estimating the transformation mapping the two frames, and obtaining a composite transformation to a global motion compensated coordinate. Sequential GMC suffers from temporal drift of frames from the accurate global coordinate, due to either error accumulation or sporadic failures of motion estimation at a few frames. We propose a temporally robust global motion compensation (TRGMC) algorithm which performs accurate and stable GMC, despite complicated and long-term camera motion. TRGMC densely connects pairs of frames, by matching local keypoints of each frame. A joint alignment of these frames is formulated as a novel keypoint-based congealing problem, where the transformation of each frame is updated iteratively, such that the spatial coordinates for the start and end points of matched keypoints are identical. Experimental results demonstrate that TRGMC has superior performance in a wide range of scenarios.
Multi-object tracking has been studied for decades. However, when it comes to tracking pedestrians in extremely crowded scenes, we are limited to only few works. This is an important problem which gives rise to several challenges. Pre-trained object detectors fail to localize targets in crowded sequences. This consequently limits the use of data-association based multi-target tracking methods which rely on the outcome of an object detector. Additionally, the small apparent target size makes it challenging to extract features to discriminate targets from their surroundings. Finally, the large number of targets greatly increases computational complexity which in turn makes it hard to extend existing multi-target tracking approaches to high-density crowd scenarios. In this paper, we propose a tracker that addresses the aforementioned problems and is capable of tracking hundreds of people efficiently. We formulate online crowd tracking as Binary Quadratic Programing. Our formulation employs target's individual information in the form of appearance and motion as well as contextual cues in the form of neighborhood motion, spatial proximity and grouping constraints, and solves detection and data association simultaneously. In order to solve the proposed quadratic optimization efficiently, where state-of art commercial quadratic programing solvers fail to find the answer in a reasonable amount of time, we propose to use the most recent version of the Modified Frank Wolfe algorithm, which takes advantage of SWAP-steps to speed up the optimization. We show that the proposed formulation can track hundreds of targets efficiently and improves state-of-art results by significant margins on eleven challenging high density crowd sequences.
Since the launch of the Microsoft Kinect, scores of RGBD datasets have been released. These have propelled advances in areas from reconstruction to gesture recognition. In this paper we explore the field, reviewing datasets across eight categories: semantics, object pose estimation, camera tracking, scene reconstruction, object tracking, human actions, faces and identification. By extracting relevant information in each category we help researchers to find appropriate data for their needs, and we consider which datasets have succeeded in driving computer vision forward and why. Finally, we examine the future of RGBD datasets. We identify key areas which are currently underexplored, and suggest that future directions may include synthetic data and dense reconstructions of static and dynamic scenes.
Deep Convolution Neural Networks (CNNs) have shown impressive performance in various vision tasks such as image classification, object detection and semantic segmentation. For object detection, particularly in still images, the performance has been significantly increased last year thanks to powerful deep networks (e.g. GoogleNet) and detection frameworks (e.g. Regions with CNN features (R-CNN)). The lately introduced ImageNet task on object detection from video (VID) brings the object detection task into the video domain, in which objects' locations at each frame are required to be annotated with bounding boxes. In this work, we introduce a complete framework for the VID task based on still-image object detection and general object tracking. Their relations and contributions in the VID task are thoroughly studied and evaluated. In addition, a temporal convolution network is proposed to incorporate temporal information to regularize the detection results and shows its effectiveness for the task.
One of the main challenges of visual object tracking comes from the arbitrary appearance of objects. Most existing algorithms try to resolve this problem as an object-specific task, i.e., the model is trained to regenerate or classify a specific object. As a result, the model need to be initialized and retrained for different objects. In this paper, we propose a more generic approach utilizing a novel two-flow convolutional neural network (named YCNN). The YCNN takes two inputs (one is object image patch, the other is search image patch), then outputs a response map which predicts how likely the object appears in a specific location. Unlike those object-specific approach, the YCNN is trained to measure the similarity between two image patches. Thus it will not be confined to any specific object. Furthermore the network can be end-to-end trained to extract both shallow and deep convolutional features which are dedicated for visual tracking. And once properly trained, the YCNN can be applied to track all kinds of objects without further training and updating. Benefiting from the once-for-all model, our algorithm is able to run at a very high speed of 45 frames-per-second. The experiments on 51 sequences also show that our algorithm achieves an outstanding performance.
We present an attention-based modular neural framework for computer vision. The framework uses a soft attention mechanism allowing models to be trained with gradient descent. It consists of three modules: a recurrent attention module controlling where to look in an image or video frame, a feature-extraction module providing a representation of what is seen, and an objective module formalizing why the model learns its attentive behavior. The attention module allows the model to focus computation on task-related information in the input. We apply the framework to several object tracking tasks and explore various design choices. We experiment with three data sets, bouncing ball, moving digits and the real-world KTH data set. The proposed Recurrent Attentive Tracking Model performs well on all three tasks and can generalize to related but previously unseen sequences from a challenging tracking data set.
Standardized benchmarks are crucial for the majority of computer vision applications. Although leaderboards and ranking tables should not be over-claimed, benchmarks often provide the most objective measure of performance and are therefore important guides for reseach. Recently, a new benchmark for Multiple Object Tracking, MOTChallenge, was launched with the goal of collecting existing and new data and creating a framework for the standardized evaluation of multiple object tracking methods. The first release of the benchmark focuses on multiple people tracking, since pedestrians are by far the most studied object in the tracking community. This paper accompanies a new release of the MOTChallenge benchmark. Unlike the initial release, all videos of MOT16 have been carefully annotated following a consistent protocol. Moreover, it not only offers a significant increase in the number of labeled boxes, but also provides multiple object classes beside pedestrians and the level of visibility for every single object of interest.
A novel probabilistic perception algorithm is presented as a real-time joint solution to data association, object tracking, and object classification for an autonomous ground vehicle in all-weather conditions. The presented algorithm extends a Rao-Blackwellized Particle Filter originally built with a particle filter for data association and a Kalman filter for multi-object tracking (Miller et al. 2011a) to now also include multiple model tracking for classification. Additionally a state-of-the-art vision detection algorithm that includes heading information for autonomous ground vehicle (AGV) applications was implemented. Cornell's AGV from the DARPA Urban Challenge was upgraded and used to experimentally examine if and how state-of-the-art vision algorithms can complement or replace lidar and radar sensors. Sensor and algorithm performance in adverse weather and lighting conditions is tested. Experimental evaluation demonstrates robust all-weather data association, tracking, and classification where camera, lidar, and radar sensors complement each other inside the joint probabilistic perception algorithm.