Speaker diarization consists of assigning speech signals to people engaged in a dialogue. An audio-visual spatiotemporal diarization model is proposed. The model is well suited for challenging scenarios that consist of several participants engaged in multi-party interaction while they move around and turn their heads towards the other participants rather than facing the cameras and the microphones. Multiple-person visual tracking is combined with multiple speech-source localization in order to tackle the speech-to-person association problem. The latter is solved within a novel audio-visual fusion method on the following grounds: binaural spectral features are first extracted from a microphone pair, then a supervised audio-visual alignment technique maps these features onto an image, and finally a semi-supervised clustering method assigns binaural spectral features to visible persons. The main advantage of this method over previous work is that it processes in a principled way speech signals uttered simultaneously by multiple persons. The diarization itself is cast into a latent-variable temporal graphical model that infers speaker identities and speech turns, based on the output of an audio-visual association process, executed at each time slice, and on the dynamics of the diarization variable itself. The proposed formulation yields an efficient exact inference procedure. A novel dataset, that contains audio-visual training data as well as a number of scenarios involving several participants engaged in formal and informal dialogue, is introduced. The proposed method is thoroughly tested and benchmarked with respect to several state-of-the art diarization algorithms.
Recently, discriminatively learned correlation filters (DCF) has drawn much attention in visual object tracking community. The success of DCF is potentially attributed to the fact that a large amount of samples are utilized to train the ridge regression model and predict the location of object. To solve the regression problem in an efficient way, these samples are all generated by circularly shifting from a search patch. However, these synthetic samples also induce some negative effects which weaken the robustness of DCF based trackers. In this paper, we propose a Convolutional Regression framework for visual tracking (CRT). Instead of learning the linear regression model in a closed form, we try to solve the regression problem by optimizing a one-channel-output convolution layer with Gradient Descent (GD). In particular, the receptive field size of the convolution layer is set to the size of object. Contrary to DCF, it is possible to incorporate all "real" samples clipped from the whole image. A critical issue of the GD approach is that most of the convolutional samples are negative and the contribution of positive samples will be suppressed. To address this problem, we propose a novel Automatic Hard Negative Mining method to eliminate easy negatives and enhance positives. Extensive experiments are conducted on a widely-used benchmark with 100 sequences. The results show that the proposed algorithm achieves outstanding performance and outperforms almost all the existing DCF based algorithms.
Visual tracking addresses the problem of identifying and localizing an unknown target in a video given the target specified by a bounding box in the first frame. In this paper, we propose a dual network to better utilize features among layers for visual tracking. It is observed that features in higher layers encode semantic context while its counterparts in lower layers are sensitive to discriminative appearance. Thus we exploit the hierarchical features in different layers of a deep model and design a dual structure to obtain better feature representation from various streams, which is rarely investigated in previous work. To highlight geometric contours of the target, we integrate the hierarchical feature maps with an edge detector as the coarse prior maps to further embed local details around the target. To leverage the robustness of our dual network, we train it with random patches measuring the similarities between the network activation and target appearance, which serves as a regularization to enforce the dual network to focus on target object. The proposed dual network is updated online in a unique manner based on the observation that the target being tracked in consecutive frames should share more similar feature representations than those in the surrounding background. It is also found that for a target object, the prior maps can help further enhance performance by passing message into the output maps of the dual network. Therefore, an independent component analysis with reference algorithm (ICA-R) is employed to extract target context using prior maps as guidance. Online tracking is conducted by maximizing the posterior estimate on the final maps with stochastic and periodic update. Quantitative and qualitative evaluations on two large-scale benchmark data sets show that the proposed algorithm performs favourably against the state-of-the-arts.
Robust visual tracking is a challenging computer vision problem, with many real-world applications. Most existing approaches employ hand-crafted appearance features, such as HOG or Color Names. Recently, deep RGB features extracted from convolutional neural networks have been successfully applied for tracking. Despite their success, these features only capture appearance information. On the other hand, motion cues provide discriminative and complementary information that can improve tracking performance. Contrary to visual tracking, deep motion features have been successfully applied for action recognition and video classification tasks. Typically, the motion features are learned by training a CNN on optical flow images extracted from large amounts of labeled videos. This paper presents an investigation of the impact of deep motion features in a tracking-by-detection framework. We further show that hand-crafted, deep RGB, and deep motion features contain complementary information. To the best of our knowledge, we are the first to propose fusing appearance information with deep motion features for visual tracking. Comprehensive experiments clearly suggest that our fusion approach with deep motion features outperforms standard methods relying on appearance information alone.
Visual tracking is a fundamental problem in computer vision. Recently, some deep-learning-based tracking algorithms have been achieving record-breaking performances. However, due to the high complexity of deep learning, most deep trackers suffer from low tracking speed, and thus are impractical in many real-world applications. Some new deep trackers with smaller network structure achieve high efficiency while at the cost of significant decrease on precision. In this paper, we propose to transfer the feature for image classification to the visual tracking domain via convolutional channel reductions. The channel reduction could be simply viewed as an additional convolutional layer with the specific task. It not only extracts useful information for object tracking but also significantly increases the tracking speed. To better accommodate the useful feature of the target in different scales, the adaptation filters are designed with different sizes. The yielded visual tracker is real-time and also illustrates the state-of-the-art accuracies in the experiment involving two well-adopted benchmarks with more than 100 test videos.
We present an approach towards robust lane tracking for assisted and autonomous driving, particularly under poor visibility. Autonomous detection of lane markers improves road safety, and purely visual tracking is desirable for widespread vehicle compatibility and reducing sensor intrusion, cost, and energy consumption. However, visual approaches are often ineffective because of a number of factors, including but not limited to occlusion, poor weather conditions, and paint wear-off. Our method, named SafeDrive, attempts to improve visual lane detection approaches in drastically degraded visual conditions without relying on additional active sensors. In scenarios where visual lane detection algorithms are unable to detect lane markers, the proposed approach uses location information of the vehicle to locate and access alternate imagery of the road and attempts detection on this secondary image. Subsequently, by using a combination of feature-based and pixel-based alignment, an estimated location of the lane marker is found in the current scene. We demonstrate the effectiveness of our system on actual driving data from locations in the United States with Google Street View as the source of alternate imagery.
One of the major challenges of model-free visual tracking problem has been the difficulty originating from the unpredictable and drastic changes in the appearance of objects we target to track. Existing methods tackle this problem by updating the appearance model on-line in order to adapt to the changes in the appearance. Despite the success of these methods however, inaccurate and erroneous updates of the appearance model result in a tracker drift. In this paper, we introduce a novel real-time visual tracking algorithm based on a template selection strategy constructed by deep reinforcement learning methods. The tracking algorithm utilizes this strategy to choose the appropriate template for tracking a given frame. The template selection strategy is self-learned by utilizing a simple policy gradient method on numerous training episodes randomly generated from a tracking benchmark dataset. Our proposed reinforcement learning framework is generally applicable to other confidence map based tracking algorithms. The experiment shows that our tracking algorithm runs in real-time speed of 43 fps and the proposed policy network effectively decides the appropriate template for successful visual tracking.
Acquiring the accurate 3-D position of a target person around a robot provides fundamental and valuable information that is applicable to a wide range of robotic tasks, including home service, navigation and entertainment. This paper presents a real-time robotic 3-D human tracking system which combines a monocular camera with an ultrasonic sensor by the extended Kalman filter (EKF). The proposed system consists of three sub-modules: monocular camera sensor tracking model, ultrasonic sensor tracking model and multi-sensor fusion. An improved visual tracking algorithm is presented to provide partial location estimation (2-D). The algorithm is designed to overcome severe occlusions, scale variation, target missing and achieve robust re-detection. The scale accuracy is further enhanced by the estimated 3-D information. An ultrasonic sensor array is employed to provide the range information from the target person to the robot and Gaussian Process Regression is used for partial location estimation (2-D). EKF is adopted to sequentially process multiple, heterogeneous measurements arriving in an asynchronous order from the vision sensor and the ultrasonic sensor separately. In the experiments, the proposed tracking system is tested in both simulation platform and actual mobile robot for various indoor and outdoor scenes. The experimental results show the superior performance of the 3-D tracking system in terms of both the accuracy and robustness.
Correlation Filters (CFs) have recently demonstrated excellent performance in terms of rapidly tracking objects under challenging photometric and geometric variations. The strength of the approach comes from its ability to efficiently learn - "on the fly" - how the object is changing over time. A fundamental drawback to CFs, however, is that the background of the object is not be modelled over time which can result in suboptimal results. In this paper we propose a Background-Aware CF that can model how both the foreground and background of the object varies over time. Our approach, like conventional CFs, is extremely computationally efficient - and extensive experiments over multiple tracking benchmarks demonstrate the superior accuracy and real-time performance of our method compared to the state-of-the-art trackers including those based on a deep learning paradigm.
Discrminative trackers, employ a classification approach to separate the target from its background. To cope with variations of the target shape and appearance, the classifier is updated online with different samples of the target and the background. Sample selection, labeling and updating the classifier is prone to various sources of errors that drift the tracker. We introduce the use of an efficient version space shrinking strategy to reduce the labeling errors and enhance its sampling strategy by measuring the uncertainty of the tracker about the samples. The proposed tracker, utilize an ensemble of classifiers that represents different hypotheses about the target, diversify them using boosting to provide a larger and more consistent coverage of the version-space and tune the classifiers' weights in voting. The proposed system adjusts the model update rate by promoting the co-training of the short-memory ensemble with a long-memory oracle. The proposed tracker outperformed state-of-the-art trackers on different sequences bearing various tracking challenges.
In this paper we introduce a fully end-to-end approach for visual tracking in videos that learns to predict the bounding box locations of a target object at every frame. An important insight is that the tracking problem can be considered as a sequential decision-making process and historical semantics encode highly relevant information for future decisions. Based on this intuition, we formulate our model as a recurrent convolutional neural network agent that interacts with a video overtime, and our model can be trained with reinforcement learning (RL) algorithms to learn good tracking policies that pay attention to continuous, inter-frame correlation and maximize tracking performance in the long run. The proposed tracking algorithm achieves state-of-the-art performance in an existing tracking benchmark and operates at frame-rates faster than real-time. To the best of our knowledge, our tracker is the first neural-network tracker that combines convolutional and recurrent networks with RL algorithms.
Discriminant Correlation Filters (DCF) based methods now become a kind of dominant approach to online object tracking. The features used in these methods, however, are either based on hand-crafted features like HoGs, or convolutional features trained independently from other tasks like image classification. In this work, we present an end-to-end lightweight network architecture, namely DCFNet, to learn the convolutional features and perform the correlation tracking process simultaneously. Specifically, we treat DCF as a special correlation filter layer added in a Siamese network, and carefully derive the backpropagation through it by defining the network output as the probability heatmap of object location. Since the derivation is still carried out in Fourier frequency domain, the efficiency property of DCF is preserved. This enables our tracker to run at more than 60 FPS during test time, while achieving a significant accuracy gain compared with KCF using HoGs. Extensive evaluations on OTB-2013, OTB-2015, and VOT2015 benchmarks demonstrate that the proposed DCFNet tracker is competitive with several state-of-the-art trackers, while being more compact and much faster.
This paper addresses the problem of online tracking and classification of multiple objects in an image sequence. Our proposed solution is to first track all objects in the scene without relying on object-specific prior knowledge, which in other systems can take the form of hand-crafted features or user-based track initialization. We then classify the tracked objects with a fast-learning image classifier that is based on a shallow convolutional neural network architecture and demonstrate that object recognition improves when this is combined with object state information from the tracking algorithm. We argue that by transferring the use of prior knowledge from the detection and tracking stages to the classification stage we can design a robust, general purpose object recognition system with the ability to detect and track a variety of object types. We describe our biologically inspired implementation, which adaptively learns the shape and motion of tracked objects, and apply it to the Neovision2 Tower benchmark data set, which contains multiple object types. An experimental evaluation demonstrates that our approach is competitive with state-of-the-art video object recognition systems that do make use of object-specific prior knowledge in detection and tracking, while providing additional practical advantages by virtue of its generality.
Convolutional neural network (CNN) has drawn increasing interest in visual tracking owing to its powerfulness in feature extraction. Most existing CNN-based trackers treat tracking as a classification problem. However, these trackers are sensitive to similar distractors because their CNN models mainly focus on inter-class classification. To address this problem, we use self-structure information of object to distinguish it from distractors. Specifically, we utilize recurrent neural network (RNN) to model object structure, and incorporate it into CNN to improve its robustness to similar distractors. Considering that convolutional layers in different levels characterize the object from different perspectives, we use multiple RNNs to model object structure in different levels respectively. Extensive experiments on three benchmarks, OTB100, TC-128 and VOT2015, show that the proposed algorithm outperforms other methods. Code is released at http://www.dabi.temple.edu/~hbling/code/SANet/SANet.html.
In this paper we address the problem of developing on-line visual tracking algorithms. We present a specialized communication protocol that serves as a bridge between a tracker implementation and utilizing application. It decouples development of algorithms and application, encouraging re-usability. The primary use case is algorithm evaluation where the protocol facilitates more complex evaluation scenarios that are used nowadays thus pushing forward the field of visual tracking. We present a reference implementation of the protocol that makes it easy to use in several popular programming languages and discuss where the protocol is already used and some usage scenarios that we envision for the future.
Robust feature representation plays significant role in visual tracking. However, it remains a challenging issue, since many factors may affect the experimental performance. The existing method which combine different features by setting them equally with the fixed weight could hardly solve the issues, due to the different statistical properties of different features across various of scenarios and attributes. In this paper, by exploiting the internal relationship among these features, we develop a robust method to construct a more stable feature representation. More specifically, we utilize a co-training paradigm to formulate the intrinsic complementary information of multi-feature template into the efficient correlation filter framework. We test our approach on challenging se- quences with illumination variation, scale variation, deformation etc. Experimental results demonstrate that the proposed method outperforms state-of-the-art methods favorably.
Recent advances in visual tracking showed that deep Convolutional Neural Networks (CNN) trained for image classification can be strong feature extractors for discriminative trackers. However, due to the drastic difference between image classification and tracking, extra treatments such as model ensemble and feature engineering must be carried out to bridge the two domains. Such procedures are either time consuming or hard to generalize well across datasets. In this paper we discovered that the internal structure of Region Proposal Network (RPN)'s top layer feature can be utilized for robust visual tracking. We showed that such property has to be unleashed by a novel loss function which simultaneously considers classification accuracy and bounding box quality. Without ensemble and any extra treatment on feature maps, our proposed method achieved state-of-the-art results on several large scale benchmarks including OTB50, OTB100 and VOT2016. We will make our code publicly available.
Discriminatively learned correlation filters (DCF) have been widely used in online visual tracking filed due to its simplicity and efficiency. These methods utilize a periodic assumption of the training samples to construct a circulant data matrix, which implicitly increases the training samples and reduces both storage and computational complexity.The periodic assumption also introduces unwanted boundary effects. Recently, Spatially Regularized Correlation Filters (SRDCF) solved this issue by introducing penalization on correlation filter coefficients depending on their spatial location. However, SRDCF's efficiency dramatically decreased due to the breaking of circulant structure. We propose Faster Spatially Regularized Discriminative Correlation Filters (FSRDCF) for tracking. The FSRDCF is constructed from Ridge Regression, the circulant structure of training samples in the spatial domain is fully used, more importantly, we further exploit the circulant structure of regularization function in the Fourier domain, which allows our problem to be solved more directly and efficiently. Experiments are conducted on three benchmark datasets: OTB-2013, OTB-2015 and VOT2016. Our approach achieves equivalent performance to the baseline tracker SRDCF on all three datasets. On OTB-2013 and OTB-2015 datasets, our approach obtains a more than twice faster running speed and a more than third times shorter start-up time than the SRDCF. For state-of-the-art comparison, our approach demonstrates superior performance compared to other non-spatial-regularization trackers.
This survey presents a deep analysis of the learning and inference capabilities in nine popular trackers. It is neither intended to study the whole literature nor is it an attempt to review all kinds of neural networks proposed for visual tracking. We focus instead on Siamese neural networks which are a promising starting point for studying the challenging problem of tracking. These networks integrate efficiently feature learning and the temporal matching and have so far shown state-of-the-art performance. In particular, the branches of Siamese networks, their layers connecting these branches, specific aspects of training and the embedding of these networks into the tracker are highlighted. Quantitative results from existing papers are compared with the conclusion that the current evaluation methodology shows problems with the reproducibility and the comparability of results. The paper proposes a novel Lisp-like formalism for a better comparison of trackers. This assumes a certain functional design and functional decomposition of trackers. The paper tries to give foundation for tracker design by a formulation of the problem based on the theory of machine learning and by the interpretation of a tracker as a decision function. The work concludes with promising lines of research and suggests future work.
Unlike standard cameras that send intensity images at a constant frame rate, event-driven cameras asynchronously report pixel-level brightness changes, offering low latency and high temporal resolution (both in the order of micro-seconds). As such, they have great potential for fast and low power vision algorithms for robots. Visual tracking, for example, is easily achieved even for very fast stimuli, as only moving objects cause brightness changes. However, cameras mounted on a moving robot are typically non-stationary and the same tracking problem becomes confounded by background clutter events due to the robot ego-motion. In this paper, we propose a method for segmenting the motion of an independently moving object for event-driven cameras. Our method detects and tracks corners in the event stream and learns the statistics of their motion as a function of the robot's joint velocities when no independently moving objects are present. During robot operation, independently moving objects are identified by discrepancies between the predicted corner velocities from ego-motion and the measured corner velocities. We validate the algorithm on data collected from the neuromorphic iCub robot. We achieve a precision of ~ 90 % and show that the method is robust to changes in speed of both the head and the target.