作者: 新星 時間: 2025-3-22 00:06 作者: overshadow 時間: 2025-3-22 03:17
Eye Movement-Based Analysis on Methodologies and Efficiency in the Process of Image Noise Evaluationistorted images for making decision on noise evaluation is rather limited. In this paper, we conducted psychophysical eye-tracking studies to deeply understand the process of image noise evaluation. We identified two different types of methodologies in the evaluation processing, speed-driven and acc作者: Additive 時間: 2025-3-22 06:29
IBDNet: Lightweight Network for On-orbit Image Blind Denoisingnerstone of image processing, image denoising exceedingly improves the image quality to contribute to subsequent works. For on-orbit image denoising, we propose an end-to-end trainable image blind denoising network, namely IBDNet. Unlike existing image denoising methods, which either have a large nu作者: Femine 時間: 2025-3-22 09:58
Aggregating Rich Deep Semantic Features for Fine-Grained Place Classificationages depends on the objects and text as well as the various semantic regions, hierarchical structure, and spatial layout. However, most recently designed fine-grained classification systems ignored this, the complex multi-level semantic structure of images associated with fine-grained classes has no作者: 效果 時間: 2025-3-22 14:50
Improving Reliability of Object Detection for Lunar Craters Using Monte Carlo Dropoutesent the uncertainty of such detection. However, a measure of uncertainty could be expressed as the variance of the prediction by using Monte Carlo Dropout Sampling (MC Dropout). Although MC Dropout has often been applied to fully connected layers in a network in recent studies, many convolutional 作者: negligence 時間: 2025-3-22 18:31
An Improved Convolutional Neural Network for Steganalysis in the Scenario of Reuse of the Stego-Keyhe steganalysis scenario of the repeated use of the stego-key is considered. Firstly, a study of the influence of the depth and width of the convolution layers on the effectiveness of classification was conducted. Next, a study on the influence of depth and width of fully connected layers on the eff作者: 帶子 時間: 2025-3-22 23:48
A New Learning-Based One Shot Detection Framework for Natural Imagesbjects well. In this paper, we propose a new framework that applies one-shot learning to object detection. During the training period, the network learns an ability from known object classes to compare the similarity of two image parts. For the image of a new category, selective search seeks proposa作者: Mystic 時間: 2025-3-23 01:40
Dense Receptive Field Network: A Backbone Network for Object Detection detection tasks. So, designing a special backbone network for detection tasks is one of the best solutions. In this paper, a backbone network named Dense Receptive Field Network (DRFNet) is proposed for object detection. DRFNet is based on Darknet-60 (our modified version of Darknet-53) and contain作者: left-ventricle 時間: 2025-3-23 05:49 作者: CLIFF 時間: 2025-3-23 10:54 作者: 強有力 時間: 2025-3-23 16:40
Severe Convective Weather Classification in Remote Sensing Images by Semantic Segmentations how to recognize severe convection weather accurately and effectively, and it is also an important issue in government climate risk management. However, most existing methods extract features from satellite data by classifying individual pixels instead of using tightly integrated spatial informati作者: 壁畫 時間: 2025-3-23 18:19
Action Recognition Based on Divide-and-Conquert of redundant information, compared with dense sampling, sparse sampling network can also achieve good results. Due to sparse sampling’s limitation of access to information, this paper mainly discusses how to further improve the learning ability of the model based on sparse sampling. We proposed a 作者: 吸引力 時間: 2025-3-24 01:12 作者: 自負的人 時間: 2025-3-24 03:15
In-Silico Staining from Bright-Field and Fluorescent Images Using Deep Learningus and costly, it damages tissue and suffers from inconsistencies. Recently deep learning approaches have been successfully applied to predict fluorescent markers from bright-field images [.,.,.]. These approaches can save costs and time and speed up the classification of tissue properties. However,作者: 記成螞蟻 時間: 2025-3-24 08:23
A Lightweight Neural Network for Hard Exudate Segmentation of Fundus Imagete, a special kind of lesion in the fundus image, is treated as the basis to evaluate the severity level of DR. Therefore, it is crucial to segment hard exudate exactly. However, the segmentation results of existing deep learning-based segmentation methods are rather coarse due to successive pooling作者: 嚴厲批評 時間: 2025-3-24 10:39
https://doi.org/10.1007/978-3-030-30508-6artificial intelligence; classification; clustering; computational linguistics; computer networks; Human-作者: craven 時間: 2025-3-24 15:07
978-3-030-30507-9Springer Nature Switzerland AG 2019作者: 真繁榮 時間: 2025-3-24 19:50 作者: 共同確定為確 時間: 2025-3-25 02:45
Lecture Notes in Computer Sciencehttp://image.papertrans.cn/b/image/162645.jpg作者: 妨礙 時間: 2025-3-25 06:31
Manfred Wick,Wulf Pinggera,Paul Lehmanninal image so that the overall performance of visualization, classification and segmentation tasks is considerably improved. Traditional techniques require manual fine-tuning of the parameters to control enhancement behavior. To date, recent Convolutional Neural Network (CNN) approaches frequently e作者: 枯萎將要 時間: 2025-3-25 10:03
Conference proceedings 19911st editionransparent object: due to refraction, the image is heavily distorted; the pinhole camera model alone can not be used and a distortion correction step is required. By directly modeling the geometry of the refractive media, we build the image generation process by tracing individual light rays from th作者: 嚴厲批評 時間: 2025-3-25 14:45
Manfred Wick,Wulf Pinggera,Paul Lehmannistorted images for making decision on noise evaluation is rather limited. In this paper, we conducted psychophysical eye-tracking studies to deeply understand the process of image noise evaluation. We identified two different types of methodologies in the evaluation processing, speed-driven and acc作者: cancer 時間: 2025-3-25 17:41 作者: 談判 時間: 2025-3-25 22:28
Conference proceedings 19942nd editionages depends on the objects and text as well as the various semantic regions, hierarchical structure, and spatial layout. However, most recently designed fine-grained classification systems ignored this, the complex multi-level semantic structure of images associated with fine-grained classes has no作者: 特別容易碎 時間: 2025-3-26 01:08
Manfred Wick,Wulf Pinggera,Paul Lehmannesent the uncertainty of such detection. However, a measure of uncertainty could be expressed as the variance of the prediction by using Monte Carlo Dropout Sampling (MC Dropout). Although MC Dropout has often been applied to fully connected layers in a network in recent studies, many convolutional 作者: Enthralling 時間: 2025-3-26 06:14 作者: 向下五度才偏 時間: 2025-3-26 09:16 作者: 地殼 時間: 2025-3-26 16:36
https://doi.org/10.1007/978-3-7091-4435-0 detection tasks. So, designing a special backbone network for detection tasks is one of the best solutions. In this paper, a backbone network named Dense Receptive Field Network (DRFNet) is proposed for object detection. DRFNet is based on Darknet-60 (our modified version of Darknet-53) and contain作者: refine 時間: 2025-3-26 20:28 作者: Gossamer 時間: 2025-3-26 22:28
Ferro- and Antiferroelectricityfor image transformation: U-Net (based on CNNs) and U-ReNet (partially based on CNNs and RNNs). In this work, we propose a novel U-ReNet which is almost entirely RNN based. We compare U-Net, U-ReNet (partially RNN), and our U-ReNet (almost entirely RNN based) in two datasets based on MNIST. The task作者: neutralize 時間: 2025-3-27 05:09
https://doi.org/10.1007/978-3-540-49604-5s how to recognize severe convection weather accurately and effectively, and it is also an important issue in government climate risk management. However, most existing methods extract features from satellite data by classifying individual pixels instead of using tightly integrated spatial informati作者: Arteriography 時間: 2025-3-27 05:24
Classification of Ferroalloy Processes,t of redundant information, compared with dense sampling, sparse sampling network can also achieve good results. Due to sparse sampling’s limitation of access to information, this paper mainly discusses how to further improve the learning ability of the model based on sparse sampling. We proposed a 作者: 凹室 時間: 2025-3-27 10:37 作者: 開始發(fā)作 時間: 2025-3-27 16:20 作者: 侵略 時間: 2025-3-27 19:09
Ferroelectric Domains: Some Recent Advances,te, a special kind of lesion in the fundus image, is treated as the basis to evaluate the severity level of DR. Therefore, it is crucial to segment hard exudate exactly. However, the segmentation results of existing deep learning-based segmentation methods are rather coarse due to successive pooling作者: Anthrp 時間: 2025-3-27 23:53 作者: cornucopia 時間: 2025-3-28 04:12 作者: 感情脆弱 時間: 2025-3-28 09:01 作者: 訓誡 時間: 2025-3-28 13:03
Manfred Wick,Wulf Pinggera,Paul Lehmanner these steps, we can obtain a temporary result. Based on this result and some proposals related to it, we refine the proposals through the intersection. Then we conduct second-round detection with new proposals and improve the accuracy. Experiments on different datasets demonstrate that our method is effective and has a certain transferability.作者: 鞠躬 時間: 2025-3-28 14:49
https://doi.org/10.1007/978-3-7091-4435-0h ScratchDet for fast evaluation. Moreover, as a pre-trained model on ImageNet, DRFNet is also tested with SSD. All the experiments show that DRFNet is an effective and efficient backbone network for object detection.作者: poliosis 時間: 2025-3-28 22:01
Phase Transitions in Thin Films,square estimation. We integrate the adaptive feature channel weighting scheme into two state-of-the-art handcrafted DCF based trackers, and evaluate them on two benchmarks: OTB2013 and VOT2016, respectively. The experimental results demonstrate its accuracy and efficiency when compared with some state-of-the-art handcrafted DCF based trackers.作者: molest 時間: 2025-3-29 00:50
Ferroelectric Domains: Some Recent Advances,s show that our network achieves superior performance with the fewest parameters and the fastest speed compared with baseline methods on the IDRiD dataset. Specially, with 1/20 parameters and 1/3 inference time, our method is over 10% higher than DeepLab v3+ in terms of F1-score on the IDRiD dataset. The source code of LWENet is available at ..作者: Fortify 時間: 2025-3-29 05:55
Eye Movement-Based Analysis on Methodologies and Efficiency in the Process of Image Noise Evaluationspatial entropy analysis on eye movement data, a quantitative measure is obtained to show significant correlation with the decision-making efficiency of evaluation processing, which is characterized by evaluation time and decision error. As a result, the new measure may be used as a proxy definition for this decision-making efficiency.作者: Charlatan 時間: 2025-3-29 08:59 作者: deficiency 時間: 2025-3-29 11:26 作者: osteoclasts 時間: 2025-3-29 16:57
A New Learning-Based One Shot Detection Framework for Natural Imageser these steps, we can obtain a temporary result. Based on this result and some proposals related to it, we refine the proposals through the intersection. Then we conduct second-round detection with new proposals and improve the accuracy. Experiments on different datasets demonstrate that our method is effective and has a certain transferability.作者: 條約 時間: 2025-3-29 22:35 作者: ventilate 時間: 2025-3-30 03:30
An Adaptive Feature Channel Weighting Scheme for Correlation Trackingsquare estimation. We integrate the adaptive feature channel weighting scheme into two state-of-the-art handcrafted DCF based trackers, and evaluate them on two benchmarks: OTB2013 and VOT2016, respectively. The experimental results demonstrate its accuracy and efficiency when compared with some state-of-the-art handcrafted DCF based trackers.作者: Hiatal-Hernia 時間: 2025-3-30 07:15 作者: COMA 時間: 2025-3-30 10:30
Conference proceedings 2019tworks, ICANN 2019, held in Munich, Germany, in September 2019.?The total of 277 full papers and 43 short papers presented in these proceedings was carefully reviewed and selected from 494 submissions. They were organized in 5 volumes focusing on theoretical neural computation; deep learning; image 作者: brachial-plexus 時間: 2025-3-30 14:22 作者: NAUT 時間: 2025-3-30 17:27 作者: AMOR 時間: 2025-3-30 21:23 作者: oblique 時間: 2025-3-31 01:20
Classification of Ferroalloy Processes,model based on divide-and-conquer, which use a threshold . to determine whether action data require sparse sampling or dense local sampling for learning. Finally, our approach obtains the state-the-of-art performance on the datasets of HMDB51 (72.4%) and UCF101 (95.3%).作者: 監(jiān)禁 時間: 2025-3-31 05:53
Comparison Between U-Net and U-ReNet Models in OCR Tasks is to transform text lines of overlapping digits to text lines of separated digits. Our model reaches the best performance in one dataset and comparable results in the other dataset. Additionally, the proposed U-ReNet with RNN upsampling has fewer parameters than U-Net and is more robust to translation transformation.作者: PARA 時間: 2025-3-31 11:29 作者: 政府 時間: 2025-3-31 15:50
0302-9743 Neural Networks, ICANN 2019, held in Munich, Germany, in September 2019.?The total of 277 full papers and 43 short papers presented in these proceedings was carefully reviewed and selected from 494 submissions. They were organized in 5 volumes focusing on theoretical neural computation; deep learni作者: IDEAS 時間: 2025-3-31 19:51
