We present a method for detecting objects in images using a single deep neural network. Our approach, named SSD, discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per feature map location. At prediction time, the network generates scores for the presence of each object category in each default box and produces adjustments to the box to better match the object shape. Additionally, the network combines predictions from multiple feature maps with different resolutions to naturally handle objects of various sizes. Our SSD model is simple relative to methods that require object proposals because it completely eliminates proposal generation and subsequent pixel or feature resampling stage and encapsulates all computation in a single network. This makes SSD easy to train and straightforward to integrate into systems that require a detection component. Experimental results on the PASCAL VOC, MS COCO, and ILSVRC datasets confirm that SSD has comparable accuracy to methods that utilize an additional object proposal step and is much faster, while providing a unified framework for both training and inference. Compared to other single stage methods, SSD has much better accuracy, even with a smaller input image size. For $300\times 300$ input, SSD achieves 72.1% mAP on VOC2007 test at 58 FPS on a Nvidia Titan X and for $500\times 500$ input, SSD achieves 75.1% mAP, outperforming a comparable state of the art Faster R-CNN model. Code is available at https://github.com/weiliu89/caffe/tree/ssd .
In this paper, we propose the first higher frame rate video dataset (called Need for Speed - NfS) and benchmark for visual object tracking. The dataset consists of 100 videos (380K frames) captured with now commonly available higher frame rate (240 FPS) cameras from real world scenarios. All frames are annotated with axis aligned bounding boxes and all sequences are manually labelled with nine visual attributes - such as occlusion, fast motion, background clutter, etc. Our benchmark provides an extensive evaluation of many recent and state-of-the-art trackers on higher frame rate sequences. We ranked each of these trackers according to their tracking accuracy and real-time performance. One of our surprising conclusions is that at higher frame rates, simple trackers such as correlation filters outperform complex methods based on deep networks. This suggests that for practical applications (such as in robotics or embedded vision), one needs to carefully tradeoff bandwidth constraints associated with higher frame rate acquisition, computational costs of real-time analysis, and the required application accuracy. Our dataset and benchmark allows for the first time (to our knowledge) systematic exploration of such issues, and will be made available to allow for further research in this space.
Object-to-camera motion produces a variety of apparent motion patterns that significantly affect performance of short-term visual trackers. Despite being crucial for designing robust trackers, their influence is poorly explored in standard benchmarks due to weakly defined, biased and overlapping attribute annotations. In this paper we propose to go beyond pre-recorded benchmarks with post-hoc annotations by presenting an approach that utilizes omnidirectional videos to generate realistic, consistently annotated, short-term tracking scenarios with exactly parameterized motion patterns. We have created an evaluation system, constructed a fully annotated dataset of omnidirectional videos and the generators for typical motion patterns. We provide an in-depth analysis of major tracking paradigms which is complementary to the standard benchmarks and confirms the expressiveness of our evaluation approach.
Recently, the Kernelized Correlation Filters tracker (KCF) achieved competitive performance and robustness in visual object tracking. On the other hand, visual trackers are not typically used in multiple object tracking. In this paper, we investigate how a robust visual tracker like KCF can improve multiple object tracking. Since KCF is a fast tracker, many can be used in parallel and still result in fast tracking. We build a multiple object tracking system based on KCF and background subtraction. Background subtraction is applied to extract moving objects and get their scale and size in combination with KCF outputs, while KCF is used for data association and to handle fragmentation and occlusion problems. As a result, KCF and background subtraction help each other to take tracking decision at every frame. Sometimes KCF outputs are the most trustworthy (e.g. during occlusion), while in some other case, it is the background subtraction outputs. To validate the effectiveness of our system, the algorithm is demonstrated on four urban video recordings from a standard dataset. Results show that our method is competitive with state-of-the-art trackers even if we use a much simpler data association step.
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.
We propose a technique that propagates information forward through video data. The method is conceptually simple and can be applied to tasks that require the propagation of structured information, such as semantic labels, based on video content. We propose a 'Video Propagation Network' that processes video frames in an adaptive manner. The model is applied online: it propagates information forward without the need to access future frames. In particular we combine two components, a temporal bilateral network for dense and video adaptive filtering, followed by a spatial network to refine features and increased flexibility. We present experiments on video object segmentation and semantic video segmentation and show increased performance comparing to the best previous task-specific methods, while having favorable runtime. Additionally we demonstrate our approach on an example regression task of color propagation in a grayscale video.
In recent years, Discriminative Correlation Filter (DCF) based methods have significantly advanced the state-of-the-art in tracking. However, in the pursuit of ever increasing tracking performance, their characteristic speed and real-time capability have gradually faded. Further, the increasingly complex models, with massive number of trainable parameters, have introduced the risk of severe over-fitting. In this work, we tackle the key causes behind the problems of computational complexity and over-fitting, with the aim of simultaneously improving both speed and performance. We revisit the core DCF formulation and introduce: (i) a factorized convolution operator, which drastically reduces the number of parameters in the model; (ii) a compact generative model of the training sample distribution, that significantly reduces memory and time complexity, while providing better diversity of samples; (iii) a conservative model update strategy with improved robustness and reduced complexity. We perform comprehensive experiments on four benchmarks: VOT2016, UAV123, OTB-2015, and TempleColor. When using expensive deep features, our tracker provides a 20-fold speedup and achieves a 13.0% relative gain in Expected Average Overlap compared to the top ranked method in the VOT2016 challenge. Moreover, our fast variant, using hand-crafted features, operates at 60 Hz on a single CPU, while obtaining 65.0% AUC on OTB-2015.
This paper tackles the task of semi-supervised video object segmentation, i.e., the separation of an object from the background in a video, given the mask of the first frame. We present One-Shot Video Object Segmentation (OSVOS), based on a fully-convolutional neural network architecture that is able to successively transfer generic semantic information, learned on ImageNet, to the task of foreground segmentation, and finally to learning the appearance of a single annotated object of the test sequence (hence one-shot). Although all frames are processed independently, the results are temporally coherent and stable. We perform experiments on two annotated video segmentation databases, which show that OSVOS is fast and improves the state of the art by a significant margin (79.8% vs 68.0%).
In Computer Vision domain, moving Object Tracking considered as one of the toughest problem.As there so many factors associated like illumination of light, noise, occlusion, sudden start and stop of moving object, shading which makes tracking even harder problem not only for dynamic background but also for static background.In this paper we present a new object tracking algorithm based on Dominant points on tracked object using Quantum particle swarm optimization (QPSO) which is a new different version of PSO based on Quantum theory. The novelty in our approach is that it can be successfully applicable in variable background as well as static background and application of quantum PSO makes the algorithm runs lot faster where other basic PSO algorithm failed to do so due to heavy computation.In our approach firstly dominants points of tracked objects detected, then a group of particles form a swarm are initialized randomly over the image search space and then start searching the curvature connected between two consecutive dominant points until they satisfy fitness criteria. Obviously it is a Multi-Swarm approach as there are multiple dominant points, as they moves, the curvature moves and the curvature movement is tracked by the swarm throughout the video and eventually when the swarm reaches optimal solution , a bounding box drawn based on particles final position.Experimental results demonstrate this proposed QPSO based method work efficiently and effectively in visual object tracking in both dynamic and static environments and run time shows that it runs closely 90% faster than basic PSO.in our approach we also apply parallelism using MatLab Parfor command to show how very less number of iteration and swarm size will enable us to successfully track object.
Visual object tracking is a challenging computer vision task with numerous real-world applications. Here we propose a simple but efficient Spectral Filter Tracking (SFT)method. To characterize rotational and translation invariance of tracking targets, the candidate image region is models as a pixelwise grid graph. Instead of the conventional graph matching, we convert the tracking into a plain least square regression problem to estimate the best center coordinate of the target. But different from the holistic regression of correlation filter based methods, SFT can operate on localized surrounding regions of each pixel (i.e.,vertex) by using spectral graph filters, which thus is more robust to resist local variations and cluttered background.To bypass the eigenvalue decomposition problem of the graph Laplacian matrix L, we parameterize spectral graph filters as the polynomial of L by spectral graph theory, in which L k exactly encodes a k-hop local neighborhood of each vertex. Finally, the filter parameters (i.e., polynomial coefficients) as well as feature projecting functions are jointly integrated into the regression model.
Feature selection is playing an increasingly significant role with respect to many computer vision applications spanning from object recognition to visual object tracking. However, most of the recent solutions in feature selection are not robust across different and heterogeneous set of data. In this paper, we address this issue proposing a robust probabilistic latent graph-based feature selection algorithm that performs the ranking step while considering all the possible subsets of features, as paths on a graph, bypassing the combinatorial problem analytically. An appealing characteristic of the approach is that it aims to discover an abstraction behind low-level sensory data, that is, relevancy. Relevancy is modelled as a latent variable in a PLSA-inspired generative process that allows the investigation of the importance of a feature when injected into an arbitrary set of cues. The proposed method has been tested on ten diverse benchmarks, and compared against eleven state of the art feature selection methods. Results show that the proposed approach attains the highest performance levels across many different scenarios and difficulties, thereby confirming its strong robustness while setting a new state of the art in feature selection domain.
Visual tracking is intrinsically a temporal problem. Discriminative Correlation Filters (DCF) have demonstrated excellent performance for high-speed generic visual object tracking. Built upon their seminal work, there has been a plethora of recent improvements relying on convolutional neural network (CNN) pretrained on ImageNet as a feature extractor for visual tracking. However, most of their works relying on ad hoc analysis to design the weights for different layers either using boosting or hedging techniques as an ensemble tracker. In this paper, we go beyond the conventional DCF framework and propose a Kernalised Multi-resolution Convnet (KMC) formulation that utilises hierarchical response maps to directly output the target movement. When directly deployed the learnt network to predict the unseen challenging UAV tracking dataset without any weight adjustment, the proposed model consistently achieves excellent tracking performance. Moreover, the transfered multi-reslution CNN renders it possible to be integrated into the RNN temporal learning framework, therefore opening the door on the end-to-end temporal deep learning (TDL) for visual tracking.
We tackle the task of semi-supervised video object segmentation, i.e. segmenting the pixels belonging to an object in the video using the ground truth pixel mask for the first frame. We build on the recently introduced one-shot video object segmentation (OSVOS) approach which uses a pretrained network and fine-tunes it on the first frame. While achieving impressive performance, at test time OSVOS uses the fine-tuned network in unchanged form and is not able to adapt to large changes in object appearance. To overcome this limitation, we propose Online Adaptive Video Object Segmentation (OnAVOS) which updates the network online using training examples selected based on the confidence of the network and the spatial configuration. Additionally, we add a pretraining step based on objectness, which is learned on PASCAL. Our experiments show that both extensions are highly effective and improve the state of the art on DAVIS to an intersection-over-union score of 85.7%.
Tracking-by-detection algorithms are widely used for visual tracking, where the problem is treated as a classification task where an object model is updated over time using online learning techniques. In challenging conditions where an object undergoes deformation or scale variations, the update step is prone to include background information in the model appearance or to lack the ability to estimate the scale change, which degrades the performance of the classifier. In this paper, we incorporate a Patch-based Adaptive Weighting with Segmentation and Scale (PAWSS) tracking framework that tackles both the scale and background problems. A simple but effective colour-based segmentation model is used to suppress background information and multi-scale samples are extracted to enrich the training pool, which allows the tracker to handle both incremental and abrupt scale variations between frames. Experimentally, we evaluate our approach on the online tracking benchmark (OTB) dataset and Visual Object Tracking (VOT) challenge datasets. The results show that our approach outperforms recent state-of-the-art trackers, and it especially improves the successful rate score on the OTB dataset, while on the VOT datasets, PAWSS ranks among the top trackers while operating at real-time frame rates.
We propose a novel video object segmentation algorithm based on pixel-level matching using Convolutional Neural Networks (CNN). Our network aims to distinguish the target area from the background on the basis of the pixel-level similarity between two object units. The proposed network represents a target object using features from different depth layers in order to take advantage of both the spatial details and the category-level semantic information. Furthermore, we propose a feature compression technique that drastically reduces the memory requirements while maintaining the capability of feature representation. Two-stage training (pre-training and fine-tuning) allows our network to handle any target object regardless of its category (even if the object's type does not belong to the pre-training data) or of variations in its appearance through a video sequence. Experiments on large datasets demonstrate the effectiveness of our model - against related methods - in terms of accuracy, speed, and stability. Finally, we introduce the transferability of our network to different domains, such as the infrared data domain.
Visual object tracking is a fundamental and time-critical vision task. Recent years have seen many shallow tracking methods based on real-time pixel-based correlation filters, as well as deep methods that have top performance but need a high-end GPU. In this paper, we learn to improve the speed of deep trackers without losing accuracy. Our fundamental insight is to take an adaptive approach, where easy frames are processed with cheap features (such as pixel values), while challenging frames are processed with invariant but expensive deep features. We formulate the adaptive tracking problem as a decision-making process, and learn an agent to decide whether to locate objects with high confidence on an early layer, or continue processing subsequent layers of a network. This significantly reduces the feed-forward cost for easy frames with distinct or slow-moving objects. We train the agent offline in a reinforcement learning fashion, and further demonstrate that learning all deep layers (so as to provide good features for adaptive tracking) can lead to near real-time average tracking speed of 23 fps on a single CPU while achieving state-of-the-art performance. Perhaps most tellingly, our approach provides a 100X speedup for almost 50% of the time, indicating the power of an adaptive approach.
This paper proposes an end-to-end trainable network, SegFlow, for simultaneously predicting pixel-wise object segmentation and optical flow in videos. The proposed SegFlow has two branches where useful information of object segmentation and optical flow is propagated bidirectionally in a unified framework. The segmentation branch is based on a fully convolutional network, which has been proved effective in image segmentation task, and the optical flow branch takes advantage of the FlowNet model. The unified framework is trained iteratively offline to learn a generic notion, and fine-tuned online for specific objects. Extensive experiments on both the video object segmentation and optical flow datasets demonstrate that introducing optical flow improves the performance of segmentation and vice versa, against the state-of-the-art algorithms.
Visual object tracking is an active topic in the computer vision domain with applications extending over numerous fields. The main sub-tasks required to build an object tracker (e.g. object detection, feature extraction and object tracking) are computation-intensive. In addition, real-time operation of the tracker is indispensable for almost all of its applications. Therefore, complete hardware or hardware/software co-design approaches are pursued for better tracker implementations. This paper presents a literature survey of the hardware implementations of object trackers over the last two decades. Although several tracking surveys exist in literature, a survey addressing the hardware implementations of the different trackers is missing. We believe this survey would fill the gap and complete the picture with the existing surveys of how to design an efficient tracker and point out the future directions researchers can follow in this field. We highlight the lack of hardware implementations for state-of-the-art tracking algorithms as well as for enhanced classical algorithms. We also stress the need for measuring the tracking performance of the hardware-based trackers. Additionally, enough details of the hardware-based trackers need to be provided to allow reasonable comparison between the different implementations.
In this paper, we propose a novel pixel-wise visual object tracking framework that can track any anonymous object in a noisy background. The framework consists of two submodels, a global attention model and a local segmentation model. The global model generates a region of interests (ROI) that the object may lie in the new frame based on the past object segmentation maps, while the local model segments the new image in the ROI. Each model uses a LSTM structure to model the temporal dynamics of the motion and appearance, respectively. To circumvent the dependency of the training data between the two models, we use an iterative update strategy. Once the models are trained, there is no need to refine them to track specific objects, making our method efficient compared to online learning approaches. We demonstrate our real time pixel-wise object tracking framework on a challenging VOT dataset
Discriminative correlation filters (DCF) have recently shown excellent performance in visual object tracking area. In this paper, we summarize the methods of updating model filter from discriminative correlation filter (DCF) based tracking algorithms and analyzes similarities and differences among these methods. We deduce the relationship between updating coefficient in high dimension (kernel trick), updating filter in frequency domain and updating filter in spatial domain, and analyze the difference among these different ways. We also analyze the difference between the updating filter directly and updating filter's numerator (object response power) with updating filter's denominator (filter's power). The experiments about comparing different updating methods and visualizing the template filters are used to prove our derivation.