Search

Today, target detection has an indispensable application in various fields. Infrared small-target detection, as a branch of target detection, can improve the perception capability of autonomous systems, and it has good application prospects in infrared alarm, automatic driving and other fields. There are many well-established algorithms that perform well in infrared small-target detection. Nevertheless, the current algorithms cannot achieve the expected detection effect in complex environments, such as background clutter, noise inundation or very small targets. We have designed an image enhancement-based detection algorithm to solve both problems through detail enhancement and target expansion. This method first enhances the mutation information, detail and edge information of the image and then improves the contrast between the target edge and the adjacent pixels to make the target more prominent. The enhancement improves the robustness of detection with background clutter or noise-flooded scenes. Moreover, bicubic interpolation is used on the input image, and the target pixels are expanded with upsampling, which enhances the detection effectiveness for tiny targets. From the results of qualitative and quantitative experiments, the algorithm proposed in this paper outperforms the existing work on various evaluation indicators. [ABSTRACT FROM AUTHOR]

Due to the increased popularity of augmented and virtual reality experiences, the interest in capturing high-resolution real-world point clouds has never been higher. Loss of details and irregularities in point cloud geometry can occur during the capturing, processing, and compression pipeline. It is essential to address these challenges by being able to upsample a low Level-of-Detail (LoD) point cloud into a high LoD point cloud. Current upsampling methods suffer from several weaknesses in handling point cloud upsampling, especially in dense real-world photo-realistic point clouds. In this paper, we present a novel geometry upsampling technique, PU-Dense, which can process a diverse set of point clouds including synthetic mesh-based point clouds, real-world high-resolution point clouds, real-world indoor LiDAR scanned objects, as well as outdoor dynamically acquired LiDAR-based point clouds. PU-Dense employs a 3D multiscale architecture using sparse convolutional networks that hierarchically reconstruct an upsampled point cloud geometry via progressive rescaling and multiscale feature extraction. The framework employs a UNet type architecture that downscales the point cloud to a bottleneck and then upscales it to a higher level-of-detail (LoD) point cloud. PU-Dense introduces a novel Feature Extraction Unit that incorporates multiscale spatial learning by employing filters at multiple sampling rates and receptive fields. The architecture is memory efficient and is driven by a binary voxel occupancy classification loss that allows it to process high-resolution dense point clouds with millions of points during inference time. Qualitative and quantitative experimental results show that our method significantly outperforms the state-of-the-art approaches by a large margin while having much lower inference time complexity. We further test our dataset on high-resolution photo-realistic datasets. In addition, our method can handle noisy data well. We further show that our approach is memory efficient compared to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The task of point cloud upsampling aims to acquire dense and uniform point sets from sparse and irregular point sets. Although significant progress has been made with deep learning models, state-of-the-art methods require ground-truth dense point sets as the supervision, which makes them limited to be trained under synthetic paired training data and not suitable to be under real-scanned sparse data. However, it is expensive and tedious to obtain large numbers of paired sparse-dense point sets as supervision from real-scanned sparse data. To address this problem, we propose a self-supervised point cloud upsampling network, named SPU-Net, to capture the inherent upsampling patterns of points lying on the underlying object surface. Specifically, we propose a coarse-to-fine reconstruction framework, which contains two main components: point feature extraction and point feature expansion, respectively. In the point feature extraction, we integrate the self-attention module with the graph convolution network (GCN) to capture context information inside and among local regions simultaneously. In the point feature expansion, we introduce a hierarchically learnable folding strategy to generate upsampled point sets with learnable 2D grids. Moreover, to further optimize the noisy points in the generated point sets, we propose a novel self-projection optimization associated with uniform and reconstruction terms as a joint loss to facilitate the self-supervised point cloud upsampling. We conduct various experiments on both synthetic and real-scanned datasets, and the results demonstrate that we achieve comparable performances to state-of-the-art supervised methods. [ABSTRACT FROM AUTHOR]

To solve the problems of low recognition rate and high miss rate of small object detection in the current target detection task, an improved YOLOv3 algorithm based on gated channel attention mechanism (EGCA) and adaptive upsampling module is proposed. Firstly, darknet-53 is used as the backbone network for image basic feature extraction. Secondly, the adaptive upsampling module is introduced to expand the low-resolution convolution feature map, which effectively enhances the fusion effect of different scale convolution feature map. Finally, EGCA attention mechanism is added before the three scale channels output the prediction results to improve the feature expression and detection ability of the network to small objects. The improved algorithm is tested on the public data set RSOD (remote sensing object detection), the accuracy of small object detection is improved by 8.2 percentage points, which is the most significant. The average accuracy AP value reaches 56.3%, which is 7.9 percentage points higher than the original algorithm. Experimental results show that the improved algorithm has better small object detection ability than the traditional YOLOv3 algorithm and other algorithms. [ABSTRACT FROM AUTHOR]

The back projection (BP) algorithm is characterized by its high performance for multi-receiver synthetic aperture sonar (SAS). For this reason, it is usually used to evaluate the imaging performance of Fourier-domain methods. However, this algorithm suffers from a large computation load, and the imaging efficiency is seriously lowered. In order to improve the imaging performance, this paper proposes focusing the multi-receiver SAS data using the BP algorithm based on two interpolators, including the linear interpolation and nearest-neighbor interpolation. The former interpolation is used to decrease the interpolation error based on adjacent sampled data; the latter estimates the data at the desired moment by assigning the data value of the nearest sample as estimated data. Then, the imaging performance of the presented method is discussed in detail based on simulations and real-data processing. With the presented method, the imaging performance can be improved without a loss of efficiency compared to nearest-neighbor interpolation without an upsampling operation. In comparison with the traditional BP algorithm, the presented method can be used to improve the imaging efficiency without any loss of performance. [ABSTRACT FROM AUTHOR]

This study aims to improve the accuracy and efficiency of detecting defects on polished metal surfaces by developing a new detection system. A network model explicitly designed for detecting defects on polished metal surfaces and a convolutional feature concentration and activation network (CFCANet) are proposed in this paper. This model significantly enhances the recognition of tiny defects by introducing a small-target detection head, ensuring high-precision detection results. In addition, the feature concentration and activation (FCA-C2f) module proposed in this study enhances the model's sensitivity to anisotropic features, thereby improving defect detection accuracy. The content-aware reassembly of features (CARAFE) upsampling algorithm is used instead of traditional nearest-neighbour interpolation methods to effectively preserve detailed information and improve the quality and efficiency of upsampling. By optimising the lighting conditions and using composite light source illumination technology, the probabilities of missed detections and false alarms can be reduced. Combined with the CFCANet detection network, the defect detection performance of the proposed method on polished metal surfaces is effectively enhanced. To validate the effectiveness of the proposed method, a new dataset for detecting defects on polished metal surfaces, PMS-DET, was constructed in this study and validated on the NEU-DET dataset. Experimental results show that CFCANet effectively improves the defect detection accuracy on polished metal surfaces, achieving a mAP 0.5:0.95 value of 42.4 % on the PMS-DET dataset, an increase of 11.9 %. The model parameters are reduced by 6.7 %, and the detection speed is improved by 28.1 %. Compared with existing detection models, this research method demonstrates significant improvements in detection accuracy, model size and computational efficiency, especially regarding GFLOPs and detection speed, proving its potential application value in practical industrial scenarios. [ABSTRACT FROM AUTHOR]

This paper proposes a framework that enables the online generation of virtual point clouds relying only on previous camera and point clouds and current camera measurements. The continuous usage of the pipeline generating virtual LIDAR measurements makes the temporal up-sampling of point clouds possible. The only requirement of the system is a camera with a higher frame rate than the LIDAR equipped to the same vehicle, which is usually provided. The pipeline first utilizes optical flow estimations from the available camera frames. Next, optical expansion is used to upgrade it to 3D scene flow. Following that, ground plane fitting is made on the previous LIDAR point cloud. Finally, the estimated scene flow is applied to the previously measured object points to generate the new point cloud. The framework’s efficiency is proved as state-of-the-art performance is achieved on the KITTI dataset.

This paper introduces an encoder–decoder-based attentional decoder network to recognize small-size lesions in chest X-ray images. In the encoder-only network, small-size lesions disappear during the down-sampling steps or are indistinguishable in the low-resolution feature maps. To address these issues, the proposed network processes images in the encoder–decoder architecture similar to U-Net families and classifies lesions by globally pooling high-resolution feature maps. However, two challenging obstacles prohibit U-Net families from being extended to classification: (1) the up-sampling procedure consumes considerable resources, and (2) there needs to be an effective pooling approach for the high-resolution feature maps. Therefore, the proposed network employs a lightweight attentional decoder and harmonic magnitude transform. The attentional decoder up-samples the given features with the low-resolution features as the key and value while the high-resolution features as the query. Since multi-scaled features interact, up-sampled features embody global context at a high resolution, maintaining pathological locality. In addition, harmonic magnitude transform is devised for pooling high-resolution feature maps in the frequency domain. We borrow the shift theorem of the Fourier transform to preserve the translation invariant property and further reduce the parameters of the pooling layer by an efficient embedding strategy. The proposed network achieves state-of-the-art classification performance on the three public chest X-ray datasets, such as NIH, CheXpert, and MIMIC-CXR. In conclusion, the proposed efficient encoder–decoder network recognizes small-size lesions well in chest X-ray images by efficiently up-sampling feature maps through an attentional decoder and processing high-resolution feature maps with harmonic magnitude transform. We open-source our implementation at https://github.com/Lab-LVM/ADNet. • We propose the AD-Net that utilizes a new encoder–decoder classification architecture. • We provide a harmonic magnitude transform that efficiently combines feature maps in a translation invariant manner, using minimal additional resources. • We achieve state-of-the-art performance on three widely-used chest X-ray datasets. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Xinyang Normal University Natural Science Edition is the property of Journal of Xinyang Normal University Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Gatherings of thousands to millions of people frequently occur for an enormous variety of educational, social, sporting, and political events, and automated counting of these high-density crowds is useful for safety, management, and measuring significance of an event. In this work, we show that the regularly accepted labeling scheme of crowd density maps for training deep neural networks may not be the most effective one. We propose an alternative inverse k-nearest neighbor (i k NN) map mechanism that, even when used directly in existing state-of-the-art network structures, shows superior performance. We also provide new network architecture mechanisms that we demonstrate in our own MUD-i k NN network architecture, which uses multi-scale drop-in replacement upsampling via transposed convolutions to take full advantage of the provided i k NN labeling. This upsampling combined with the i k NN maps further improves crowd counting accuracy. We further analyze several variations of the i k NN labeling mechanism, which apply transformations on the k NN measure before generating the map, in order to consider the impact of camera perspective views, image resolutions, and the changing rates of the mapping functions. To alleviate the effects of crowd density changes in each image, we also introduce an attenuation mechanism in the i k NN mapping. Experimentally, we show that inverse square root k NN map variation (iR k NN) provides the best performance. Discussions are provided on computational complexity, label resolutions, the gains in mapping and upsampling, and details of critical cases such as various crowd counts, uneven crowd densities, and crowd occlusions. [ABSTRACT FROM AUTHOR]

Given sparse point clouds, this paper develops a perturbation learning-based point cloud upsampling method to generate uniform, clean, and dense point clouds. We build a simple yet efficient neural network framework including feature extraction, perturbation learning, and coordinate reconstruction operations. In the feature extraction task, shallow-and-wide dense connections are applied to present the latent geometric information. Subsequently, the extracted features are expanded for perturbation learning. According to the theory of the differential geometry of surfaces, the position of an upsampled point can be approximated by its projection on a tangent plane in a sufficiently small neighborhood around the point. Inspired by this, we propose learning a 2D perturbation through multilayer perceptrons (MLPs) to estimate the coordinate shift from the input point to the upsampled point. Then, we concatenate the 2D perturbation with the extracted features for residual learning to fine-tune the coordinate shift. Finally, the coordinate reconstruction step transforms all the high-level features into an upsampled and consolidated point cloud. To enable the learning-based point cloud upsampling process above, we collect a large-scale point cloud dataset that contains 36000 pairs of training sets. The entire network size is only 5.01 MB, which is much smaller than the requirements of state-of-the-art point cloud upsampling models. Qualitative and quantitative evaluation results show that our proposed scheme outperforms most existing methods in terms of the Chamfer distance (CD), Hausdorff distance (HD), Jensen-Shannon divergence (JSD), and uniformity. In addition, our method is applied to real scan data, and its robustness is confirmed. [ABSTRACT FROM AUTHOR]

Aiming at the issues of decreasing spatial resolution and feature loss caused by pooling operation in depth CNN-based hyperspectral image classification algorithm,a bilateral fusion block network （DFBN）composed of bilateral fusion blocks was designed.The upper structure of bilateral fusion block was constituted by 1×1 convolution and hyperlink,which was used to transfer local spatial characteristics.The lower structure was constituted by pooling layer,convolutional layer,deconvolution layer and upsampling to enhance the characteristics of efficient discrimination.Experimental results on three benchmark hyperspectral image data sets illustrate that the model is superior to other similar classification models.

Abstract Training neural networks with small and imbalanced datasets often leads to overfitting and disregard of the minority class. For predictive toxicology, however, models with a good balance between sensitivity and specificity are needed. In this paper we introduce conformational oversampling as a means to balance and oversample datasets for prediction of toxicity. Conformational oversampling enhances a dataset by generation of multiple conformations of a molecule. These conformations can be used to balance, as well as oversample a dataset, thereby increasing the dataset size without the need of artificial samples. We show that conformational oversampling facilitates training of neural networks and provides state-of-the-art results on the Tox21 dataset.

The flare stack is a flare gas combustion facility used to ensure the safe production of petrochemical enterprises. However, in petrochemical enterprises, cameras are far away from the flare stacks, which causes the collected flare soot pictures to be too blurry to be analyzed further. To that end, we devise an Upsampling Flare Soot Pictures Director (UFSPD) based on the global perception and local details. First, we get the global perception map and the local details map of a low resolution flare soot picture via guided filtering and edge extraction, respectively. Then, the upsampling method is used to upsample the above two maps to get the corresponding high-resolution picture. Finally, the two pictures in the second step are fused at the pixel-level to get a satisfactory flare soot picture. Experiments show that the proposed UFSPD model is more competitive than other state-of-the-art models in some image quality evaluation indexes, and it also helps to improve the detection rate of flare soot.

Joint 3D face reconstruction and dense alignment based on a single image has always been a challenge in computer vision. Recently, some work achieves these two goals simultaneously by adopting the position map to represent a 3D face. In this paper, we design a novel upsampling block, denoted as shortcut-upsampling block, to construct the position map. We first utilize the residual blocks to extract and downsample the feature maps from the 2D face image, and then utilize the proposed shortcut-upsampling blocks to convert the feature maps to the corresponding position map. Shortcut-upsampling block allows our model to be only 121M and achieve excellent performance on 3D face reconstruction and dense alignment. Compared with other deep learning methods based on 3D Morphable Model(3DMM), our model is the fastest when reconstructing a 3D face. In addition, through introducing a dynamic weight loss function, our model can converge the loss to a lower value and obtain a better performance on 3D face reconstruction.

The flare, which aims to burn all kinds of waste gas produced by petrochemical enterprises, is an important facility for plants to keep safe and prevent harming the environment. However, when the exhaust gas is not sufficiently burned, black smoke will be produced in the flare system and endanger air quality and human health. So, as a crucial task, reducing the emission of flare black smoke is attracting a growing amount of research attention. In a typical flare smoke reduction system, a high-resolution smoke image is greatly beneficial to smoke recognition and analysis. But the smoke image usually encounters the low-resolution problem. Accordingly, in this paper we propose a super-resolution method specific for smoke images, which is called smoke images upsampling method (SIUM). Considering the texture and edge characteristics of the smoke images, the proposed SIUM learns a mapping between the low-resolution images and the associated high-resolution images. Experimental results demonstrate that our proposed SIUM is prominently superior to relevant state-of-the-art technologies when applied to upsample low-resolution smoke images.

Today, target detection has an indispensable application in various fields. Infrared small-target detection, as a branch of target detection, can improve the perception capability of autonomous systems, and it has good application prospects in infrared alarm, automatic driving and other fields. There are many well-established algorithms that perform well in infrared small-target detection. Nevertheless, the current algorithms cannot achieve the expected detection effect in complex environments, such as background clutter, noise inundation or very small targets. We have designed an image enhancement-based detection algorithm to solve both problems through detail enhancement and target expansion. This method first enhances the mutation information, detail and edge information of the image and then improves the contrast between the target edge and the adjacent pixels to make the target more prominent. The enhancement improves the robustness of detection with background clutter or noise-flooded scenes. Moreover, bicubic interpolation is used on the input image, and the target pixels are expanded with upsampling, which enhances the detection effectiveness for tiny targets. From the results of qualitative and quantitative experiments, the algorithm proposed in this paper outperforms the existing work on various evaluation indicators.

The back projection (BP) algorithm is characterized by its high performance for multi-receiver synthetic aperture sonar (SAS). For this reason, it is usually used to evaluate the imaging performance of Fourier-domain methods. However, this algorithm suffers from a large computation load, and the imaging efficiency is seriously lowered. In order to improve the imaging performance, this paper proposes focusing the multi-receiver SAS data using the BP algorithm based on two interpolators, including the linear interpolation and nearest-neighbor interpolation. The former interpolation is used to decrease the interpolation error based on adjacent sampled data; the latter estimates the data at the desired moment by assigning the data value of the nearest sample as estimated data. Then, the imaging performance of the presented method is discussed in detail based on simulations and real-data processing. With the presented method, the imaging performance can be improved without a loss of efficiency compared to nearest-neighbor interpolation without an upsampling operation. In comparison with the traditional BP algorithm, the presented method can be used to improve the imaging efficiency without any loss of performance.

In this paper, an efficient super-resolution (SR) method based on deep convolutional neural network (CNN) is proposed, namely gradual upsampling network (GUN). Recent CNN-based SR methods often preliminarily magnify the low-resolution (LR) input to high-resolution (HR) input and then reconstruct the HR input, or directly reconstruct the LR input and then recover the HR result at the last layer. The proposed GUN utilizes a gradual process instead of these two commonly used frameworks. The GUN consists of an input layer, multiple upsampling and convolutional layers, and an output layer. By means of the gradual process, the proposed network can simplify the direct SR problem to multistep easier upsampling tasks with very small magnification factor in each step. Furthermore, a gradual training strategy is presented for the GUN. In the proposed training process, an initial network can be easily trained with edgelike samples, and then, the weights are gradually tuned with more complex samples. The GUN can recover fine and vivid results and is easy to be trained. The experimental results on several image sets demonstrate the effectiveness of the proposed network.

Copyright of Journal on Communication / Tongxin Xuebao is the property of Journal on Communications Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

A novel automatic image segmentation technique in magnetic resonance imaging (MRI) based on di‐phase midway convolution and deconvolution network is proposed. It consists of three convolutional and deconvolutional blocks for downsampling and upsampling layers respectively. In first block, each input slice is separately convolved using two paths with 3 × 3 and 7 × 7 kernels to produce different feature maps. Then the mean value of these feature maps is processed into upcoming blocks in downsampling and upsampling layers. This processed outcome is classified and segmented using softmax classification. Further, the volume, probability density distribution of tumor, and normal tissue regions are calculated using tissue‐type mapping technique. This method is extensively tested with BRATS 2012, BRATS 2013, and BRATS 2018 data sets. Our experimental results achieved higher dice similarity coefficient values of 24.3%, 27.5%, and 3.4%, respectively, for these three data sets when compared to the state‐of‐art brain tumor segmentation methods. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel upsampling framework to enhance the spatial resolution of the depth image. In our framework, the upscaling of a low-resolution depth image is guided by a corresponding intensity images; we formulate it as a cost aggregation problem with the guided filter. However, the guided filter does not make full use of the information of the depth image. Since depth images have quite sparse gradients, it inspires us to regularize the gradients for improving depth upscaling results. Statistics show a special property of depth images, that is, there is a non-ignorable part of pixels whose horizontal or vertical derivatives are equal to ± 1 . Based on this special property, we propose a low gradient regularization method which reduces the penalty for horizontal or vertical derivative ± 1 , and well describes the statistics of the depth image gradients. Then, we present a solution to the low gradient minimization problem based on threshold shrinkage. Finally, the proposed low gradient regularization is integrated with the guided filter into the depth image upsampling method. Experimental results demonstrate the effectiveness of our proposed approach both qualitatively and quantitatively compared with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Training neural networks with small and imbalanced datasets often leads to overfitting and disregard of the minority class. For predictive toxicology, however, models with a good balance between sensitivity and specificity are needed. In this paper we introduce conformational oversampling as a means to balance and oversample datasets for prediction of toxicity. Conformational oversampling enhances a dataset by generation of multiple conformations of a molecule. These conformations can be used to balance, as well as oversample a dataset, thereby increasing the dataset size without the need of artificial samples. We show that conformational oversampling facilitates training of neural networks and provides state-of-the-art results on the Tox21 dataset. [ABSTRACT FROM AUTHOR]

In recent years, many deep-network-based super-resolution techniques have been proposed and have achieved impressive results for 2 ${\times }$ and higher magnification factors. However, lower magnification factors encountered in some industrial applications have not received special attention, such as 720P-to-1080P (1.5 ${\times }$ magnification). Compared to traditional 2 ${\times }$ or higher magnification factors, these lower magnifications are much simpler, but reconstructions of high-definition images are time-consuming and computationally complex. Hence, in this paper, a fast image upsampling method is designed specifically for industrial applications at low magnification. In the proposed method, edge and nonedge areas are first distinguished and then reconstructed via different fast approaches. For the edge area, a local edge pattern encoding-based method is presented to recover sharp edges. For the nonedge area, a global iterative reconstruction with texture constraint is utilized. Moreover, some acceleration strategies are also presented to further reduce the complexity. The experimental results demonstrate that the proposed method can obtain performance comparable to that of some state-of-the-art methods for 720P-to-1080P magnification, but the computational cost is much lower. [ABSTRACT FROM AUTHOR]

The Web images and videos are often downsampled and compressed to save the bandwidth and storage. Hence, the low-quality and low-resolution Web images/videos cannot match the high-definition display devices nowadays. Unfortunately, traditional image super-resolution (SR) methods are not very robust to compression artifacts. In this paper, we propose an efficient joint SR and deblocking method based on simple three-step-process, which consists of a block-matching and 3D filtering process, a local binary encoding process, and a mapping reconstruction process. Furthermore, the cascade framework and an extra post-processing are also presented for large magnification factors. Experimental results on real-world Web images with obvious compression artifacts demonstrate that the proposed method can reproduce clear and sharp SR results, and effectively remove the unnatural artifacts at the same time.

In this paper, we investigate video quality enhancement using computation offloading to the mobile cloud computing (MCC) environment. Our objective is to reduce the computational complexity required to covert a low-resolution video to high-resolution video while minimizing computation at the mobile client and additional communication costs. To do so, we propose an energy-efficient computation offloading framework for video streaming services in a MCC over the fifth generation (5G) cellular networks. In the proposed framework, the mobile client offloads the computational burden for the video enhancement to the cloud, which renders the side information needed to enhance video without requiring much computation by the client. The cloud detects edges from the upsampled ultra-high-resolution video (UHD) and then compresses and transmits them as side information with the original low-resolution video (e.g., full HD). Finally, the mobile client decodes the received content and integrates the SI and original content, which produces a high-quality video. In our extensive simulation experiments, we observed that the amount of computation needed to construct a UHD video in the client is 50%-60% lower than that required to decode UHD video compressed by legacy video encoding algorithms. Moreover, the bandwidth required to transmit a full HD video and its side information is around 70% lower than that required for a normal UHD video. The subjective quality of the enhanced UHD is similar to that of the original UHD video even though the client pays lower communication costs with reduced computing power. [ABSTRACT FROM AUTHOR]

Depth maps are often obtained simultaneously alongside color images, but their resolution is much lower than that of the respective color images. This paper proposes a confidence-based joint guided filter for depth map upsampling using corresponding color information. Many previous studies have been conducted to incorporate color images into the estimation of upsampling filter coefficients. The scheme proposed in this paper instead focuses on the order of the filter for unreliable pixel refinement, and the order is calculated using the confidence map derived from the shape of unreliable regions, depth map, and color pixel values. In terms of quantitative and qualitative measurements, the proposed method demonstrates superior performance over current state-of-the-art algorithms. • It proposes the confidence-based upsampling scheme for depth map. • The confidence values are computed from the shape of edge and color-depth correlation. • Then, the confidence values adjust the filter parameters for the guided filter instead of the geometric proximity. • The guided filtering is iteratively applied to refine the unreliable regions. • The proposed method greatly improves the qualitative and quantitative performance. [ABSTRACT FROM AUTHOR]

High quality upsampling of sparse 3D point clouds is critically useful for a wide range of geometric operations such as reconstruction, rendering, meshing, and analysis. In this paper, we propose a data-driven algorithm that enables an upsampling of 3D point clouds without the need for hard-coded rules. Our approach uses a deep network with Chamfer distance as the loss function, capable of learning the latent features in point clouds belonging to different object categories. We evaluate our algorithm across different amplification factors, with upsampling learned and performed on objects belonging to the same category as well as different categories. We also explore the desirable characteristics of input point clouds as a function of the distribution of the point samples. Finally, we demonstrate the performance of our algorithm in single-category training versus multi-category training scenarios. The final proposed model is compared against a baseline, optimization-based upsampling method. The results indicate that our algorithm is capable of generating more accurate upsamplings with less Chamfer loss. [ABSTRACT FROM AUTHOR]

Location-based social media have been extensively utilized in the concept of “social sensing” to exploit dynamic information about human activities, yet joint uses of social sensing and remote sensing images are underdeveloped at present. In this paper, the close relationship between the number of Twitter users and brightness of nighttime lights (NTL) over the contiguous United States is calculated and geotagged tweets are then used to upsample a stable light image for 2013. An associated outcome of the upsampling process is the solution of two major problems existing in the NTL image, pixel saturation, and blooming effects. Compared with the original stable light image, digital number (DN) values of the upsampled stable light image have larger correlation coefficients with gridded population (0.47 versus 0.09) and DN values of the new generation NTL image product (0.56 versus 0.52), i.e., the Visible Infrared Imaging Radiometer Suite day/night band image composite. In addition, total personal incomes of states are disaggregated to each pixel in proportion to the DN value of the pixel in the NTL images and then aggregate by counties. Personal incomes distributed by the upsampled NTL image are closer to the official demographic data than those distributed by the original stable light image. All of these results explore the potential of geotagged tweets to improve the quality of NTL images for more accurately estimating or mapping socioeconomic factors. [ABSTRACT FROM AUTHOR]