In this paper we purpose several methods for creating 2D projection image from a 3D scan such as from CT or MR scans, and then look at different types of 2.5D projections with a growing window sizes and see its effect...
详细信息
Remote sensing technology is now being used with unprecedented high resolution, making important contributions in practical applications such as urban development, construction planning and disaster prediction. Howeve...
详细信息
Plug-and-Play (PnP) methods are efficient iterative algorithms for solving ill-posed image inverse problems. PnP methods are obtained by using deep Gaussian denoisers instead of the proximal operator or the gradient-d...
详细信息
ISBN:
(纸本)9781713899921
Plug-and-Play (PnP) methods are efficient iterative algorithms for solving ill-posed image inverse problems. PnP methods are obtained by using deep Gaussian denoisers instead of the proximal operator or the gradient-descent step within proximal algorithms. Current PnP schemes rely on data-fidelity terms that have either Lipschitz gradients or closed-form proximal operators, which is not applicable to Poisson inverse problems. Based on the observation that the Gaussian noise is not the adequate noise model in this setting, we propose to generalize PnP using the Bregman Proximal Gradient (BPG) method. BPG replaces the Euclidean distance with a Bregman divergence that can better capture the smoothness properties of the problem. We introduce the Bregman Score Denoiser specifically parametrized and trained for the new Bregman geometry and prove that it corresponds to the proximal operator of a nonconvex potential. We propose two PnP algorithms based on the Bregman Score Denoiser for solving Poisson inverse problems. Extending the convergence results of BPG in the nonconvex settings, we show that the proposed methods converge, targeting stationary points of an explicit global functional. Experimental evaluations conducted on various Poisson inverse problems validate the convergence results and showcase effective restoration performance.
3D segmentation of nuclei images is a fundamental task for many biological studies. Despite the rapid advances of large-volume 3D imaging acquisition methods and the emergence of sophisticated algorithms to segment th...
详细信息
ISBN:
(纸本)9781713899921
3D segmentation of nuclei images is a fundamental task for many biological studies. Despite the rapid advances of large-volume 3D imaging acquisition methods and the emergence of sophisticated algorithms to segment the nuclei in recent years, a benchmark with all cells completely annotated is still missing, making it hard to accurately assess and further improve the performance of the algorithms. The existing nuclei segmentation benchmarks either worked on 2D only or annotated a small number of 3D cells, perhaps due to the high cost of 3D annotation for large-scale data. To fulfill the critical need, we constructed NIS3D, a 3D, high cell density, large-volume, and completely annotated Nuclei image Segmentation benchmark, assisted by our newly designed semi-automatic annotation software. NIS3D provides more than 22,000 cells across multiple most-used species in this area. Each cell is labeled by three independent annotators, so we can measure the variability of each annotation. A confidence score is computed for each cell, allowing more nuanced testing and performance comparison. A comprehensive review on the methods of segmenting 3D dense nuclei was conducted. The benchmark was used to evaluate the performance of several selected state-of-the-art segmentation algorithms. The best of current methods is still far away from human-level accuracy, corroborating the necessity of generating such a benchmark. The testing results also demonstrated the strength and weakness of each method and pointed out the directions of further methodological development. The dataset can be downloaded here https://***/yu-lab-vt/NIS3D.
The quantity of data in any field of activity is increasing at an exponential rate on an annual basis. images are no exception to this rule. In such circumstances, the importance of image analysis and p...
详细信息
Modeling and producing lifelike clothed human images has attracted researchers' attention from different areas for decades, with the complexity from highly articulated and structured content. Rendering algorithms ...
The image Recognition Search Engine is one of the modern advances in artificial intelligence and image analysis and processing. This advanced system surpasses typical functionalities in image identification as it capt...
详细信息
image restoration and image enhancement are fundamental tasks in computer vision and imageprocessing. image restoration aims to recover the original information from a degraded or corrupted image, while image enhance...
详细信息
Early diagnosis of diseases has become the major focus of researchers today. Machine Learning (ML) and Deep Learning (DL) algorithms have provided a much-needed boost to this field to make early diagnosis possible. Wi...
详细信息
ISBN:
(纸本)9789819916474;9789819916481
Early diagnosis of diseases has become the major focus of researchers today. Machine Learning (ML) and Deep Learning (DL) algorithms have provided a much-needed boost to this field to make early diagnosis possible. With the help of sufficient data and computation, DL algorithms can be developed to detect diseases with high precision. Even with this uplift, gathering high-quality and quantity data for medical studies is very difficult. Although implementing Federated Learning (FL) solves data availability issues, its high variance makes it incompatible with Medical Diagnosis. In this paper, we present an FL approach for diagnosing pneumonia in patients that is easy to use and provides state-of-the-art results. Before training, a combination of pre-processing steps is performed that significantly increase the performance of the FL architecture.
The proceedings contain 72 papers. The topics discussed include: classification of degradation level of pineapple leaf spot disease using MobileNetV3;multi-processing depth map estimation for single-view 3D image gene...
ISBN:
(纸本)9798400717055
The proceedings contain 72 papers. The topics discussed include: classification of degradation level of pineapple leaf spot disease using MobileNetV3;multi-processing depth map estimation for single-view 3D image generation;artistic outpainting through adaptive image-to-text and text-to-image generation;visual presentation learning with contrastive learning for radiology report generation;data-driven methodology for bike route identification to enhance urban cycling infrastructure in metropolitan manila;person re-identification from video surveillance systems using artificial intelligence methods;assessing the security posture of mobile automated fingerprint identification systems (MABIS): a comprehensive analysis for enhanced resilience;and facemasks detection with different orientation using deep learning algorithms.
暂无评论