This paper proposes to learn reliable dense correspondence from videos in a self-supervised manner. Our learning process integrates two highly related tasks: tracking large image regions \emph{and} establishing fine-grained pixel-level associations between consecutive video frames. We exploit the synergy between both tasks through a shared inter-frame affinity matrix, which simultaneously models transitions between video frames at both the region- and pixel-levels. While region-level localization helps reduce ambiguities in fine-grained matching by narrowing down search regions; fine-grained matching provides bottom-up features to facilitate region-level localization. Our method outperforms the state-of-the-art self-supervised methods on a variety of visual correspondence tasks, including video-object and part-segmentation propagation, keypoint tracking, and object tracking. Our self-supervised method even surpasses the fully-supervised affinity feature representation obtained from a ResNet-18 pre-trained on the ImageNet.
Object detection and tracking are challenging tasks for resource-constrained embedded systems. While these tasks are among the most compute-intensive tasks from the artificial intelligence domain, they are only allowed to use limited computation and memory resources on embedded devices. In the meanwhile, such resource-constrained implementations are often required to satisfy additional demanding requirements such as real-time response, high-throughput performance, and reliable inference accuracy. To overcome these challenges, we propose SkyNet, a hardware-efficient neural network to deliver the state-of-the-art detection accuracy and speed for embedded systems. Instead of following the common top-down flow for compact DNN (Deep Neural Network) design, SkyNet provides a bottom-up DNN design approach with comprehensive understanding of the hardware constraints at the very beginning to deliver hardware-efficient DNNs. The effectiveness of SkyNet is demonstrated by winning the competitive System Design Contest for low power object detection in the 56th IEEE/ACM Design Automation Conference (DAC-SDC), where our SkyNet significantly outperforms all other 100+ competitors: it delivers 0.731 Intersection over Union (IoU) and 67.33 frames per second (FPS) on a TX2 embedded GPU; and 0.716 IoU and 25.05 FPS on an Ultra96 embedded FPGA. The evaluation of SkyNet is also extended to GOT-10K, a recent large-scale high-diversity benchmark for generic object tracking in the wild. For state-of-the-art object trackers SiamRPN++ and SiamMask, where ResNet-50 is employed as the backbone, implementations using our SkyNet as the backbone DNN are 1.60X and 1.73X faster with better or similar accuracy when running on a 1080Ti GPU, and 37.20X smaller in terms of parameter size for significantly better memory and storage footprint.
In recent years, multi object tracking (MOT) problem has drawn attention to it and has been studied in various research areas. However, some of the challenging problems including time dependent cardinality, unordered measurement set, and object labeling remain unclear. In this paper, we propose robust nonparametric methods to model the state prior for MOT problem. These models are shown to be more flexible and robust compared to existing methods. In particular, the overall approach estimates time dependent object cardinality, provides object labeling, and identifies object associated measurements. Moreover, our proposed framework dynamically contends with the birth/death and survival of the objects through dependent nonparametric processes. We present Inference algorithms that demonstrate the utility of the dependent nonparametric models for tracking. We employ Monte Carlo sampling methods to demonstrate the proposed algorithms efficiently learn the trajectory of objects from noisy measurements. The computational results display the performance of the proposed algorithms and comparison not only between one another, but also between proposed algorithms and labeled multi Bernoulli tracker.
The fully-convolutional siamese network based on template matching has shown great potentials in visual tracking. During testing, the template is fixed with the initial target feature and the performance totally relies on the general matching ability of the siamese network. However, this manner cannot capture the temporal variations of targets or background clutter. In this work, we propose a novel gradient-guided network to exploit the discriminative information in gradients and update the template in the siamese network through feed-forward and backward operations. Our algorithm performs feed-forward and backward operations to exploit the discriminative informaiton in gradients and capture the core attention of the target. To be specific, the algorithm can utilize the information from the gradient to update the template in the current frame. In addition, a template generalization training method is proposed to better use gradient information and avoid overfitting. To our knowledge, this work is the first attempt to exploit the information in the gradient for template update in siamese-based trackers. Extensive experiments on recent benchmarks demonstrate that our method achieves better performance than other state-of-the-art trackers.
Multi-sensor perception is crucial to ensure the reliability and accuracy in autonomous driving system, while multi-object tracking (MOT) improves that by tracing sequential movement of dynamic objects. Most current approaches for multi-sensor multi-object tracking are either lack of reliability by tightly relying on a single input source (e.g., center camera), or not accurate enough by fusing the results from multiple sensors in post processing without fully exploiting the inherent information. In this study, we design a generic sensor-agnostic multi-modality MOT framework (mmMOT), where each modality (i.e., sensors) is capable of performing its role independently to preserve reliability, and further improving its accuracy through a novel multi-modality fusion module. Our mmMOT can be trained in an end-to-end manner, enables joint optimization for the base feature extractor of each modality and an adjacency estimator for cross modality. Our mmMOT also makes the first attempt to encode deep representation of point cloud in data association process in MOT. We conduct extensive experiments to evaluate the effectiveness of the proposed framework on the challenging KITTI benchmark and report state-of-the-art performance. Code and models are available at https://github.com/ZwwWayne/mmMOT.
The problem of visual object tracking has traditionally been handled by variant tracking paradigms, either learning a model of the object's appearance exclusively online or matching the object with the target in an offline-trained embedding space. Despite the recent success, each method agonizes over its intrinsic constraint. The online-only approaches suffer from a lack of generalization of the model they learn thus are inferior in target regression, while the offline-only approaches (e.g., convolutional siamese trackers) lack the target-specific context information thus are not discriminative enough to handle distractors, and robust enough to deformation. Therefore, we propose an online module with an attention mechanism for offline siamese networks to extract target-specific features under L2 error. We further propose a filter update strategy adaptive to treacherous background noises for discriminative learning, and a template update strategy to handle large target deformations for robust learning. Effectiveness can be validated in the consistent improvement over three siamese baselines: SiamFC, SiamRPN++, and SiamMask. Beyond that, our model based on SiamRPN++ obtains the best results over six popular tracking benchmarks and can operate beyond real-time.
A lot a research is focused on object detection and it has achieved significant advances with deep learning techniques in recent years. Inspite of the existing research, these algorithms are not usually optimal for dealing with sequences or images captured by drone-based platforms, due to various challenges such as view point change, scales, density of object distribution and occlusion. In this paper, we develop a model for detection of objects in drone images using the VisDrone2019 DET dataset. Using the RetinaNet model as our base, we modify the anchor scales to better handle the detection of dense distribution and small size of the objects. We explicitly model the channel interdependencies by using "Squeeze-and-Excitation" (SE) blocks that adaptively recalibrates channel-wise feature responses. This helps to bring significant improvements in performance at a slight additional computational cost. Using this architecture for object detection, we build a custom DeepSORT network for object detection on the VisDrone2019 MOT dataset by training a custom Deep Association network for the algorithm.
Plant Phenomics can be used to monitor the health and the growth of plants. Computer vision applications like stereo reconstruction, image retrieval, object tracking, and object recognition play an important role in imaging based plant phenotyping. This paper offers a comparative evaluation of some popular 3D correspondence grouping algorithms, motivated by the important role that they can play in tasks such as model creation, plant recognition and identifying plant parts. Another contribution of this paper is the extension of 2D maximum likelihood matching to 3D Maximum Likelihood Estimation Sample Consensus (MLEASAC). MLESAC is efficient and is computationally less intense than 3D random sample consensus (RANSAC). We test these algorithms on 3D point clouds of plants along with two standard benchmarks addressing shape retrieval and point cloud registration scenarios. The performance is evaluated in terms of precision and recall.
Object tracking has been studied for decades, but most of the existing works are focused on the short-term tracking. For a long sequence, the object is often fully occluded or out of view for a long time, and existing short-term object tracking algorithms often lose the target, and it is difficult to re-catch the target even if it reappears again. In this paper a novel long-term object tracking algorithm flow_MDNet_RPN is proposed, in which a tracking result judgement module and a detection module are added to the short-term object tracking algorithm. Experiments show that the proposed long-term tracking algorithm is effective to the problem of target disappearance.
Due to better video quality and higher frame rate, the performance of multiple object tracking issues has been greatly improved in recent years. However, in real application scenarios, camera motion and noisy per frame detection results degrade the performance of trackers significantly. High-speed and high-quality multiple object trackers are still in urgent demand. In this paper, we propose a new multiple object tracker following the popular tracking-by-detection scheme. We tackle the camera motion problem with an optical flow network and utilize an auxiliary tracker to deal with the missing detection problem. Besides, we use both the appearance and motion information to improve the matching quality. The experimental results on the VisDrone-MOT dataset show that our approach can improve the performance of multiple object tracking significantly while achieving a high efficiency.
Feature detectors and descriptors have been successfully used for various computer vision tasks, such as video object tracking and content-based image retrieval. Many methods use image gradients in different stages of the detection-description pipeline to describe local image structures. Recently, some, or all, of these stages have been replaced by convolutional neural networks (CNNs), in order to increase their performance. A detector is defined as a selection problem, which makes it more challenging to implement as a CNN. They are therefore generally defined as regressors, converting input images to score maps and keypoints can be selected with non-maximum suppression. This paper discusses and compares several recent methods that use CNNs for keypoint detection. Experiments are performed both on the CNN based approaches, as well as a selection of conventional methods. In addition to qualitative measures defined on keypoints and descriptors, the bag-of-words (BoW) model is used to implement an image retrieval application, in order to determine how the methods perform in practice. The results show that each type of features are best in different contexts.
Convolutional neural networks (CNN) based tracking approaches have shown favorable performance in recent benchmarks. Nonetheless, the chosen CNN features are always pre-trained in different tasks and individual components in tracking systems are learned separately, thus the achieved tracking performance may be suboptimal. Besides, most of these trackers are not designed towards real-time applications because of their time-consuming feature extraction and complex optimization details. In this paper, we propose an end-to-end framework to learn the convolutional features and perform the tracking process simultaneously, namely, a unified convolutional tracker (UCT). Specifically, the UCT treats feature extractor and tracking process both as convolution operation and trains them jointly, which enables learned CNN features are tightly coupled with tracking process. During online tracking, an efficient model updating method is proposed by introducing peak-versus-noise ratio (PNR) criterion, and scale changes are handled efficiently by incorporating a scale branch into network. Experiments are performed on four challenging tracking datasets: OTB2013, OTB2015, VOT2015 and VOT2016. Our method achieves leading performance on these benchmarks while maintaining beyond real-time speed.
Image-based fiducial markers are useful in problems such as object tracking in cluttered or textureless environments, camera (and multi-sensor) calibration tasks, and vision-based simultaneous localization and mapping (SLAM). The state-of-the-art fiducial marker detection algorithms rely on the consistency of the ambient lighting. This paper introduces LiDARTag, a novel fiducial tag design and detection algorithm suitable for light detection and ranging (LiDAR) point clouds. The proposed method runs in real-time and can process data at 100 Hz, which is faster than the currently available LiDAR sensor frequencies. Because of the LiDAR sensors' nature, rapidly changing ambient lighting will not affect the detection of a LiDARTag; hence, the proposed fiducial marker can operate in a completely dark environment. In addition, the LiDARTag nicely complements and is compatible with existing visual fiducial markers, such as AprilTags, allowing for efficient multi-sensor fusion and calibration tasks. We further propose a concept of minimizing a fitting error between a point cloud and the marker's template to estimate the marker's pose. The proposed method achieves millimeter error in translation and a few degrees in rotation. Due to LiDAR returns' sparsity, the point cloud is lifted to a continuous function in a reproducing kernel Hilbert space where the inner product can be used to determine a marker's ID. The experimental results, verified by a motion capture system, confirm that the proposed method can reliably provide a tag's pose and unique ID code. The rejection of false positives is validated on the Google Cartographer indoor dataset and the Honda H3D outdoor dataset. All implementations are coded in C++ and are available at: https://github.com/UMich-BipedLab/LiDARTag.
Visual object tracking is a fundamental task in the field of computer vision. Recently, Siamese trackers have achieved state-of-the-art performance on recent benchmarks. However, Siamese trackers do not fully utilize semantic and objectness information from pre-trained networks that have been trained on the image classification task. Furthermore, the pre-trained Siamese architecture is sparsely activated by the category label which leads to unnecessary calculations and overfitting. In this paper, we propose to learn a Domain-Aware, that is fully utilizing semantic and objectness information while producing a class-agnostic using a ridge regression network. Moreover, to reduce the sparsity problem, we solve the ridge regression problem with a differentiable weighted-dynamic loss function. Our tracker, dubbed DomainSiam, improves the feature learning in the training phase and generalization capability to other domains. Extensive experiments are performed on five tracking benchmarks including OTB2013 and OTB2015 for a validation set; as well as the VOT2017, VOT2018, LaSOT, TrackingNet, and GOT10k for a testing set. DomainSiam achieves state-of-the-art performance on these benchmarks while running at 53 FPS.
Intuitively, motion blur may hurt the performance of visual object tracking. However, we lack quantitative evaluation of tracker robustness to different levels of motion blur. Meanwhile, while image deblurring methods can produce visually clearer videos for pleasing human eyes, it is unknown whether visual object tracking can benefit from image deblurring or not. In this paper, we address these two problems by constructing a Blurred Video Tracking benchmark, which contains a variety of videos with different levels of motion blurs, as well as ground truth tracking results for evaluating trackers. We extensively evaluate 23 trackers on this benchmark and observe several new interesting results. Specifically, we find that light blur may improve the performance of many trackers, but heavy blur always hurts the tracking performance. We also find that image deblurring may help to improve tracking performance on heavily blurred videos but hurt the performance on lightly blurred videos. According to these observations, we propose a new GAN based scheme to improve the tracker robustness to motion blurs. In this scheme, a finetuned discriminator is used as an adaptive assessor to selectively deblur frames during the tracking process. We use this scheme to successfully improve the accuracy and robustness of 6 trackers.
Binarized convolutional neural networks (BCNNs) are widely used to improve memory and computation efficiency of deep convolutional neural networks (DCNNs) for mobile and AI chips based applications. However, current BCNNs are not able to fully explore their corresponding full-precision models, causing a significant performance gap between them. In this paper, we propose rectified binary convolutional networks (RBCNs), towards optimized BCNNs, by combining full-precision kernels and feature maps to rectify the binarization process in a unified framework. In particular, we use a GAN to train the 1-bit binary network with the guidance of its corresponding full-precision model, which significantly improves the performance of BCNNs. The rectified convolutional layers are generic and flexible, and can be easily incorporated into existing DCNNs such as WideResNets and ResNets. Extensive experiments demonstrate the superior performance of the proposed RBCNs over state-of-the-art BCNNs. In particular, our method shows strong generalization on the object tracking task.
Formulating the multi object tracking problem as a network flow optimization problem is a popular choice. In this paper an efficient way of learning the weights of such a network is presented. It separates the problem into one embedding of feasible solutions into a one dimensional feature space and one optimization problem. The embedding can be learned using standard SGD type optimization without relying on an additional optimizations within each step. Training data is produced by performing small perturbations of ground truth tracks and representing them using generalized graph differences, which is an efficient way introduced to represent the difference between two graphs. The proposed method is evaluated on DukeMTMCT with competitive results.
RGB-Thermal object tracking attempt to locate target object using complementary visual and thermal infrared data. Existing RGB-T trackers fuse different modalities by robust feature representation learning or adaptive modal weighting. However, how to integrate dual attention mechanism for visual tracking is still a subject that has not been studied yet. In this paper, we propose two visual attention mechanisms for robust RGB-T object tracking. Specifically, the local attention is implemented by exploiting the common visual attention of RGB and thermal data to train deep classifiers. We also introduce the global attention, which is a multi-modal target-driven attention estimation network. It can provide global proposals for the classifier together with local proposals extracted from previous tracking result. Extensive experiments on two RGB-T benchmark datasets validated the effectiveness of our proposed algorithm.
Multiple-Object Tracking (MOT) is of crucial importance for applications such as retail video analytics and video surveillance. Object detectors are often the computational bottleneck of modern MOT systems, limiting their use for real-time applications. In this paper, we address this issue by leveraging on an ensemble of detectors, each running every f frames. We measured the performance of our system in the MOT16 benchmark. The proposed model surpassed other online entries of the MOT16 challenge in speed, while maintaining an acceptable accuracy.
Due to implicitly introduced periodic shifting of limited searching area, visual object tracking using correlation filters often has to confront undesired boundary effect. As boundary effect severely degrade the quality of object model, it has made it a challenging task for unmanned aerial vehicles (UAV) to perform robust and accurate object following. Traditional hand-crafted features are also not precise and robust enough to describe the object in the viewing point of UAV. In this work, a novel tracker with online enhanced background learning is specifically proposed to tackle boundary effects. Real background samples are densely extracted to learn as well as update correlation filters. Spatial penalization is introduced to offset the noise introduced by exceedingly more background information so that a more accurate appearance model can be established. Meanwhile, convolutional features are extracted to provide a more comprehensive representation of the object. In order to mitigate changes of objects' appearances, multi-frame technique is applied to learn an ideal response map and verify the generated one in each frame. Exhaustive experiments were conducted on 100 challenging UAV image sequences and the proposed tracker has achieved state-of-the-art performance.