Multiple object tracking (MOT) in urban traffic aims to produce the trajectories of the different road users that move across the field of view with different directions and speeds and that can have varying appearances and sizes. Occlusions and interactions among the different objects are expected and common due to the nature of urban road traffic. In this work, a tracking framework employing classification label information from a deep learning detection approach is used for associating the different objects, in addition to object position and appearances. We want to investigate the performance of a modern multiclass object detector for the MOT task in traffic scenes. Results show that the object labels improve tracking performance, but that the output of object detectors are not always reliable.
This paper addresses the problem of reassembling images from disjointed fragments. More specifically, given an unordered set of fragments, we aim at reassembling one or several possibly incomplete images. The main contributions of this work are: 1) several deep neural architectures to predict the relative position of image fragments that outperform the previous state of the art; 2) casting the reassembly problem into the shortest path in a graph problem for which we provide several construction algorithms depending on available information; 3) a new dataset of images taken from the Metropolitan Museum of Art (MET) dedicated to image reassembly for which we provide a clear setup and a strong baseline.
Face hallucination is a technique that reconstruct high-resolution (HR) faces from low-resolution (LR) faces, by using the prior knowledge learned from HR/LR face pairs. Most state-of-the-arts leverage position-patch prior knowledge of human face to estimate the optimal representation coefficients for each image patch. However, they focus only the position information and usually ignore the context information of image patch. In addition, when they are confronted with misalignment or the Small Sample Size (SSS) problem, the hallucination performance is very poor. To this end, this study incorporates the contextual information of image patch and proposes a powerful and efficient context-patch based face hallucination approach, namely Thresholding Locality-constrained Representation and Reproducing learning (TLcR-RL). Under the context-patch based framework, we advance a thresholding based representation method to enhance the reconstruction accuracy and reduce the computational complexity. To further improve the performance of the proposed algorithm, we propose a promotion strategy called reproducing learning. By adding the estimated HR face to the training set, which can simulates the case that the HR version of the input LR face is present in the training set, thus iteratively enhancing the final hallucination result. Experiments demonstrate that the proposed TLcR-RL method achieves a substantial increase in the hallucinated results, both subjectively and objectively. Additionally, the proposed framework is more robust to face misalignment and the SSS problem, and its hallucinated HR face is still very good when the LR test face is from the real-world. The MATLAB source code is available at https://github.com/junjun-jiang/TLcR-RL
Scene flow describes the 3D position as well as the 3D motion of each pixel in an image. Such algorithms are the basis for many state-of-the-art autonomous or automated driving functions. For verification and training large amounts of ground truth data is required, which is not available for real data. In this paper, we demonstrate a technology to create synthetic data with dense and precise scene flow ground truth.
We propose a self-supervised learning method to jointly reason about spatial and temporal context for video recognition. Recent self-supervised approaches have used spatial context [9, 34] as well as temporal coherency [32] but a combination of the two requires extensive preprocessing such as tracking objects through millions of video frames [59] or computing optical flow to determine frame regions with high motion [30]. We propose to combine spatial and temporal context in one self-supervised framework without any heavy preprocessing. We divide multiple video frames into grids of patches and train a network to solve jigsaw puzzles on these patches from multiple frames. So the network is trained to correctly identify the position of a patch within a video frame as well as the position of a patch over time. We also propose a novel permutation strategy that outperforms random permutations while significantly reducing computational and memory constraints. We use our trained network for transfer learning tasks such as video activity recognition and demonstrate the strength of our approach on two benchmark video action recognition datasets without using a single frame from these datasets for unsupervised pretraining of our proposed video jigsaw network.
In this paper, we propose an novel implementation of a simultaneous localization and mapping (SLAM) system based on a monocular camera from an unmanned aerial vehicle (UAV) using Depth prediction performed with Capsule Networks (CapsNet), which possess improvements over the drawbacks of the more widely-used Convolutional Neural Networks (CNN). An Extended Kalman Filter will assist in estimating the position of the UAV so that we are able to update the belief for the environment. Results will be evaluated on a benchmark dataset to portray the accuracy of our intended approach.
Vehicle detection is a significant and challenging task in aerial remote sensing applications. Most existing methods detect vehicles with regular rectangle boxes and fail to offer the orientation of vehicles. However, the orientation information is crucial for several practical applications, such as the trajectory and motion estimation of vehicles. In this paper, we propose a novel deep network, called rotatable region-based residual network (R$^3$-Net), to detect multi-oriented vehicles in aerial images and videos. More specially, R$^3$-Net is utilized to generate rotatable rectangular target boxes in a half coordinate system. First, we use a rotatable region proposal network (R-RPN) to generate rotatable region of interests (R-RoIs) from feature maps produced by a deep convolutional neural network. Here, a proposed batch averaging rotatable anchor (BAR anchor) strategy is applied to initialize the shape of vehicle candidates. Next, we propose a rotatable detection network (R-DN) for the final classification and regression of the R-RoIs. In R-DN, a novel rotatable position sensitive pooling (R-PS pooling) is designed to keep the position and orientation information simultaneously while downsampling the feature maps of R-RoIs. In our model, R-RPN and R-DN can be trained jointly. We test our network on two open vehicle detection image datasets, namely DLR 3K Munich Dataset and VEDAI Dataset, demonstrating the high precision and robustness of our method. In addition, further experiments on aerial videos show the good generalization capability of the proposed method and its potential for vehicle tracking in aerial videos. The demo video is available at https://youtu.be/xCYD-tYudN0.
We showcase a family of common failures of state-of-the art object detectors. These are obtained by replacing image sub-regions by another sub-image that contains a trained object. We call this "object transplanting". Modifying an image in this manner is shown to have a non-local impact on object detection. Slight changes in object position can affect its identity according to an object detector as well as that of other objects in the image. We provide some analysis and suggest possible reasons for the reported phenomena.
We propose a novel real-time direct monocular visual odometry for omnidirectional cameras. Our method extends direct sparse odometry (DSO) by using the unified omnidirectional model as a projection function, which can be applied to fisheye cameras with a field-of-view (FoV) well above 180 degrees. This formulation allows for using the full area of the input image even with strong distortion, while most existing visual odometry methods can only use a rectified and cropped part of it. Model parameters within an active keyframe window are jointly optimized, including the intrinsic/extrinsic camera parameters, 3D position of points, and affine brightness parameters. Thanks to the wide FoV, image overlap between frames becomes bigger and points are more spatially distributed. Our results demonstrate that our method provides increased accuracy and robustness over state-of-the-art visual odometry algorithms.
Beetle antennae search (BAS) is an efficient meta-heuristic algorithm. However, the convergent results of BAS rely heavily on the random beetle direction in every iterations. More specifically, different random seeds may cause different optimized results. Besides, the step-size update algorithm of BAS cannot guarantee objective become smaller in iterative process. In order to solve these problems, this paper proposes Beetle Swarm Antennae Search Algorithm (BSAS) which combines swarm intelligence algorithm with feedback-based step-size update strategy. BSAS employs k beetles to find more optimal position in each moving rather than one beetle. The step-size updates only when k beetles return without better choices. Experiments are carried out on building system identification. The results reveal the efficacy of the BSAS algorithm to avoid influence of random direction of Beetle. In addition, the estimation errors decrease as the beetles number goes up.
Accurately localising object proposals is an important precondition for high detection rate for the state-of-the-art object detection frameworks. The accuracy of an object detection method has been shown highly related to the average recall (AR) of the proposals. In this work, we propose an advanced object proposal network in favour of translation-invariance for objectness classification, translation-variance for bounding box regression, large effective receptive fields for capturing global context and scale-invariance for dealing with a range of object sizes from extremely small to large. The design of the network architecture aims to be simple while being effective and with real time performance. Without bells and whistles the proposed object proposal network significantly improves the AR at 1,000 proposals by $35\%$ and $45\%$ on PASCAL VOC and COCO dataset respectively and has a fast inference time of 44.8 ms for input image size of $640^{2}$. Empirical studies have also shown that the proposed method is class-agnostic to be generalised for general object proposal.
Radiation therapy presents a need for dynamic tracking of a target tumor volume. Fiducial markers such as implanted gold seeds have been used to gate radiation delivery but the markers are invasive and gating significantly increases treatment time. Pretreatment acquisition of a respiratory correlated 4DCT allows for determination of accurate motion tracking which is useful in treatment planning. We design a patient-specific motion subspace and a deep convolutional neural network to recover anatomical positions from a single fluoroscopic projection in real-time. We use this deep network to approximate the nonlinear inverse of a diffeomorphic deformation composed with radiographic projection. This network recovers subspace coordinates to define the patient-specific deformation of the lungs from a baseline anatomic position. The geometric accuracy of the subspace deformations on real patient data is similar to accuracy attained by original image registration between individual respiratory-phase image volumes.
Archaeologists are in dire need of automated object reconstruction methods. Fragments reassembly is close to puzzle problems, which may be solved by computer vision algorithms. As they are often beaten on most image related tasks by deep learning algorithms, we study a classification method that can solve jigsaw puzzles. In this paper, we focus on classifying the relative position: given a couple of fragments, we compute their local relation (e.g. on top). We propose several enhancements over the state of the art in this domain, which is outperformed by our method by 25\%. We propose an original dataset composed of pictures from the Metropolitan Museum of Art. We propose a greedy reconstruction method based on the predicted relative positions.
Machine vision systems using convolutional neural networks (CNNs) for robotic applications are increasingly being developed. Conventional vision CNNs are driven by camera frames at constant sample rate, thus achieving a fixed latency and power consumption tradeoff. This paper describes further work on the first experiments of a closed-loop robotic system integrating a CNN together with a Dynamic and Active Pixel Vision Sensor (DAVIS) in a predator/prey scenario. The DAVIS, mounted on the predator Summit XL robot, produces frames at a fixed 15 Hz frame-rate and Dynamic Vision Sensor (DVS) histograms containing 5k ON and OFF events at a variable frame-rate ranging from 15-500 Hz depending on the robot speeds. In contrast to conventional frame-based systems, the latency and processing cost depends on the rate of change of the image. The CNN is trained offline on the 1.25h labeled dataset to recognize the position and size of the prey robot, in the field of view of the predator. During inference, combining the ten output classes of the CNN allows extracting the analog position vector of the prey relative to the predator with a mean 8.7% error in angular estimation. The system is compatible with conventional deep learning technology, but achieves a variable latency-power tradeoff that adapts automatically to the dynamics. Finally, investigations on the robustness of the algorithm, a human performance comparison and a deconvolution analysis are also explored.
This paper describes how to apply self-attention with relative positional encodings to the task of relation extraction. We propose to use the self-attention encoder layer together with an additional position-aware attention layer that takes into account positions of the query and the object in the sentence. The self-attention encoder also uses a custom implementation of relative positional encodings which allow each word in the sentence to take into account its left and right context. The evaluation of the model is done on the TACRED dataset. The proposed model relies only on attention (no recurrent or convolutional layers are used), while improving performance w.r.t. the previous state of the art.
Capturing and labeling camera images in the real world is an expensive task, whereas synthesizing labeled images in a simulation environment is easy for collecting large-scale image data. However, learning from only synthetic images may not achieve the desired performance in the real world due to a gap between synthetic and real images. We propose a method that transfers learned detection of an object position from a simulation environment to the real world. This method uses only a significantly limited dataset of real images while leveraging a large dataset of synthetic images using variational autoencoders. Additionally, the proposed method consistently performed well in different lighting conditions, in the presence of other distractor objects, and on different backgrounds. Experimental results showed that it achieved accuracy of 1.5mm to 3.5mm on average. Furthermore, we showed how the method can be used in a real-world scenario like a "pick-and-place" robotic task.
We present a novel object tracking scheme that can track rigid objects in real time. The approach uses subpixel-precise image edges to track objects with high accuracy. It can determine the object position, scale, and rotation with subpixel-precision at around 80fps. The tracker returns a reliable score for each frame and is capable of self diagnosing a tracking failure. Furthermore, the choice of the similarity measure makes the approach inherently robust against occlusion, clutter, and nonlinear illumination changes. We evaluate the method on sequences from rigid objects from the OTB-2015 and VOT2016 dataset and discuss its performance. The evaluation shows that the tracker is more accurate than state-of-the-art real-time trackers while being equally robust.
Machine learning (ML) is increasingly deployed in real world contexts, supplying actionable insights and forming the basis of automated decision-making systems. While issues resulting from biases pre-existing in training data have been at the center of the fairness debate, these systems are also affected by technical and emergent biases, which often arise as context-specific artifacts of implementation. This position paper interprets technical bias as an epistemological problem and emergent bias as a dynamical feedback phenomenon. In order to stimulate debate on how to change machine learning practice to effectively address these issues, we explore this broader view on bias, stress the need to reflect on epistemology, and point to value-sensitive design methodologies to revisit the design and implementation process of automated decision-making systems.
In many agricultural applications one wants to characterize physical properties of plants and use the measurements to predict, for example biomass and environmental influence. This process is known as phenotyping. Traditional collection of phenotypic information is labor-intensive and time-consuming. Use of imagery is becoming popular for phenotyping. In this paper, we present methods to estimate traits of sorghum plants from RBG cameras on board of an unmanned aerial vehicle (UAV). The position and orientation of the imagery together with the coordinates of sparse points along the area of interest are derived through a new triangulation method. A rectified orthophoto mosaic is then generated from the imagery. The number of leaves is estimated and a model-based method to analyze the leaf morphology for leaf segmentation is proposed. We present a statistical model to find the location of each individual sorghum plant.
Traditional image stitching algorithms use transforms such as homography to combine different views of a scene. They usually work well when the scene is planar or when the camera is only rotated, keeping its position static. This severely limits their use in real world scenarios where an unmanned aerial vehicle (UAV) potentially hovers around and flies in an enclosed area while rotating to capture a video sequence. We utilize known scene geometry along with recorded camera trajectory to create cylindrical images captured in a given environment such as a tunnel where the camera rotates around its center. The captured images of the inner surface of the given scene are combined to create a composite panoramic image that is textured onto a 3D geometrical object in Unity graphical engine to create an immersive environment for end users.