Search

Optical flow is the pattern of apparent motion of objects in a scene. The computation of optical flow is a critical component in numerous computer vision tasks such as object detection, visual object tracking, and activity recognition. Despite a lot of research, efficiently managing abrupt changes in motion remains a challenge in motion estimation. This paper proposes novel variational regularization methods to address this problem since they allow combining different mathematical concepts into a joint energy minimization framework. In this work, we incorporate concepts from signal sparsity into variational regularization for motion estimation. The proposed regularization uses robust $\ell _{1}$ norm, which promotes sparsity and handles motion discontinuities. By using this regularization, we promote the sparsity of the optical flow gradient. This sparsity helps recover a signal even with just a few measurements. We explore recovering optical flow from a limited set of linear measurements using this regularizer. Our findings show that leveraging the sparsity of the derivatives of optical flow reduces computational complexity and memory needs.

Abstract The design of acoustic metasurfaces with desirable properties is challenging due to their artificial nature and the large space of physical and geometrical parameters. Until recently, design strategies were primarily based on numerical simulations based on finite-element or finite-difference time-domain methods, which are limited in terms of computational speed or complexity. Here, we present an efficient two-stage data-driven approach for analyzing and designing membrane-type metasurface absorbers with desirable characteristics. In the first stage, a forward model consisting of a neural network is trained to map an input, comprising the membrane parameters, to the observed sound absorption spectrum. In the second stage, the learned forward model is inverted to infer the input parameters that produce the desired absorption response. The metasurface membrane parameters, which serve as input to the neural network, are estimated by minimizing a loss function between the desired absorption profile and the output of the learned forward model. Two devices are then fabricated using the estimated membrane parameters. The measured acoustic absorption responses of the fabricated devices show a very close agreement with the desired responses.

The incorporation of weather variables is crucial in developing an effective demand forecasting model because electricity demand is strongly influenced by weather conditions. The dependence of demand on weather conditions may change with time during a day. Therefore, the time stamped weather information is essential. In this paper, a multi-layer moving window approach is proposed to incorporate the significant weather variables, which are selected using Pearson and Spearman correlation techniques. The multi-layer moving window approach allows the layers to adjust their size to accommodate the weather variables based on their significance, which creates more flexibility and adaptability thereby improving the overall performance of the proposed approach. Furthermore, a recursive model is developed to forecast the demand in multi-step ahead. An electricity demand data for the state of New South Wales, Australia are acquired from the Australian Energy Market Operator and the associated results are reported in the paper. The results show that the proposed approach with dynamic incorporation of weather variables is promising for day-ahead and week-ahead load demand forecasting.

Pedestrian lane detection is a crucial task in assistive navigation for vision-impaired people. It can provide information on walkable regions, help blind people stay on the pedestrian lane, and assist with obstacle detection. An accurate and real-time lane detection algorithm can improve travel safety and efficiency for the visually impaired. Despite its importance, pedestrian lane detection in unstructured scenes for assistive navigation has not attracted sufficient attention in the research community. This paper aims to provide a comprehensive review and an experimental evaluation of methods that can be applied for pedestrian lane detection, thereby laying a foundation for future research in this area. Our study covers traditional and deep learning methods for pedestrian lane detection, general road detection, and general semantic segmentation. We also perform an experimental evaluation of the representative methods on a large benchmark dataset that is specifically created for pedestrian lane detection. We hope this paper can serve as an informative guide for researchers in assistive technologies, and facilitate urgently-needed research for vision-impaired people.

Depth estimation is an essential component in computer vision systems for achieving 3D scene understanding. Efficient and accurate depth map estimation has numerous applications including self-driving vehicles and virtual reality tools. This paper presents a new deep network, called D-Net, for depth estimation from a single RGB image. The proposed network can be trained end-to-end, and its structure can be customised to meet different requirements in model size, speed, and prediction accuracy. Our approach gathers strong global and local contextual features at multiple resolutions, and then transfers these to high resolutions for clearer depth maps. For the encoder backbone, D-Net can utilise many state-of-the-art models including EfficientNet, HRNet and Swin Transformer to obtain dense depth maps. The proposed D-net is designed to have minimal parameters and reduced computational complexity. Extensive evaluations on the NYUv2 and KITTI benchmark datasets show that our model is highly accurate across multiple backbones, and it achieves state-of-the-art performance on both benchmarks when combined with the Swin Transformer and HRNets.

Detecting electricity price spikes is crucial as it helps market participants to gain confidence and formulate appropriate strategy to maximize their benefits. In this paper, a multi-feature based approach with the incorporation of variable thresholds is developed to detect electricity price spikes in the national electricity market of Australia. The variable thresholds, which are determined using a weighted sliding window average and an adjusted standard deviation, help to segregate spikes from normal price variations. Also, significant features are extracted from the market after thoroughly analyzing the underlying causes resulting into the price spikes. These features are employed as inputs to a support vector machine to classify electricity prices as spikes or non-spikes. A case study is conducted using a dataset acquired from the state of New South Wales, Australia. The results show that the proposed method can successfully detect the price spikes with high accuracy and confidence.

With the advances in sonar imaging technology, sonar imagery has increasingly been used for oceanographic studies in civilian and military applications. High-resolution imaging sonars can be mounted on various survey platforms, typically autonomous underwater vehicles, which provide enhanced speed and improved data quality with long-range support. This paper addresses the automatic detection of mine-like objects using sonar images. The proposed Gabor-based detector is designed as a feature pyramid network with a small number of trainable weights. Our approach combines both semantically weak and strong features to handle mine-like objects at multiple scales effectively. For feature extraction, we introduce a parameterized Gabor layer which improves the generalization capability and computational efficiency. The steerable Gabor filtering modules are embedded within the cascaded layers to enhance the scale and orientation decomposition of images. The entire deep Gabor neural network is trained in an end-to-end manner from input sonar images with annotated mine-like objects. An extensive experimental evaluation on a real sonar dataset shows that the proposed method achieves competitive performance compared to the existing approaches.

Image restoration from noisy observations is an inverse problem. Total variation (TV) is widely used to regularize this problem. TV preserves object boundaries better than a quadratic regularizer; however, it performs poor in low-textured image regions because it generates undesirable staircase artefacts. Furthermore, TV can preserve sharp horizontal and vertical edges; however, it causes the unnecessary smoothing of edges at an angle other than 0o or 90o. This problem arises because TV minimizes the gradient magnitude. Therefore, to preserve sharp boundaries, the design of an efficient variational regularizer is crucial. This paper presents a novel regularizer for the denoising of multi-channel vector valued image. The proposed regularizer uses horizontal, vertical as well as diagonal derivatives, and imposes the intensity continuity of partial image derivatives at each pixel of the underlying image. Experiments reveal that the proposed regularizer preserves edges and object boundaries better than TV based regularizers. This regularizer is also able to reduce undesirable staircase artefacts produced by TV in flat image regions.

Abstract ■■■ In stereo matching applications, local cost aggregation techniques are usually preferred over global methods due to their speed and ease of implementation. Local methods make implicit smoothness assumptions by aggregating costs within a finite window; however, cost aggregation is a time-consuming process. Furthermore, most existing local methods are based on pixel intensity values, and hence are not efficient with feature vectors used in wide-baseline stereo matching. In this paper, a new cost aggregation method is proposed, where a Per-Column Cost matrix is combined with a feature-vector-based weighting strategy to achieve both matching accuracy and computational efficiency. Here, the proposed cost aggregation method is applied with the DAISY feature descriptor for wide-baseline stereo matching; however, this method can also be applied to a fast growing number of stereo matching techniques that are based on feature descriptors. A performance comparison with several benchmark local cost aggregation approaches is presented, along with a thorough analysis of the time and storage complexity of the proposed method.

Echo planar imaging (EPI), a fast magnetic resonance imaging technique, is a powerful tool in functional neuroimaging studies. However, susceptibility artifacts, which cause misinterpretations of brain functions, are unavoidable distortions in EPI. This paper proposes an end-to-end deep learning framework, named TS-Net, for susceptibility artifact correction (SAC) in a pair of 3D EPI images with reversed phase-encoding directions. The proposed TS-Net comprises a deep convolutional network to predict a displacement field in three dimensions to overcome the limitation of existing methods, which only estimate the displacement field along the dominant-distortion direction. In the training phase, anatomical T1-weighted images are leveraged to regularize the correction, but they are not required during the inference phase to make TS-Net more flexible for general use. The experimental results show that TS-Net achieves favorable accuracy and speed trade-off when compared with the state-of-the-art SAC methods, i.e., TOPUP, TISAC, and S-Net. The fast inference speed (less than a second) of TS-Net makes real-time SAC during EPI image acquisition feasible and accelerates the medical image-processing pipelines.

Source positioning using hybrid angle-of-arrival (AOA) estimation and received signal strength indicator (RSSI) is attractive because no synchronization is required among unknown nodes and anchors. Conventionally, hybrid AOA/RSSI localization combines the same number of these measurements to estimate the agents’ locations. However, since AOA estimation requires anchors to be equipped with large antenna arrays and complicated signal processing, this conventional combination makes the wireless sensor network (WSN) complicated. This paper proposes an unbalanced integration of the two measurements, called 1AOA/nRSSI, to simplify the WSN. Instead of using many anchors with large antenna arrays, the proposed method only requires one master anchor to provide one AOA estimation, while other anchors are simple single-antenna transceivers. By simply transforming the 1AOA/1RSSI information into two corresponding virtual anchors, the problem of integrating one AOA and N RSSI measurements is solved using the least square and subspace methods. The solutions are then evaluated to characterize the impact of angular and distance measurement errors. Simulation results show that the proposed network achieves the same level of precision as in a fully hybrid nAOA/nRSSI network with a slightly higher number of simple anchors.

An image classification technique, which has recently been introduced for visual pattern recognition, is successfully applied for human gait classification based on radar Doppler signatures depicted in the time-frequency domain. The proposed method has three processing stages. The first two stages are designed to extract Doppler features that can effectively characterize human motion based on the nature of arm swings, and the third stage performs classification. Three types of arm motion are considered: free-arm swings, one-arm confined swings, and no-arm swings. The last two arm motions can be indicative of a human carrying objects or a person in stressed situations. The paper discusses the different steps of the proposed method for extracting distinctive Doppler features and demonstrates their contributions to the final and desirable classification rates.