Semi-supervised learning for raindrop removal on a single image

2021 ◽  
pp. 1-9
Author(s):  
Juan Zhang ◽  
Wenbin Guo
Keyword(s):  
Supervised Learning ◽  
Experimental Results ◽  
Real Image ◽  
Single Image ◽  
The Real ◽  
Rainy Weather

This article propose s a network that is mainly used to deal with a single image polluted by raindrops in rainy weather to get a clean image without raindrops. In the existing solutions, most of the methods rely on paired images, that is, the rain image and the real image without rain in the same scene. However, in many cases, the paired images are difficult to obtain, which makes it impossible to apply the raindrop removal network in many scenarios. Therefore this article proposes a semi-supervised rain-removing network apply to unpaired images. The model contains two parts: a supervised network and an unsupervised network. After the model is trained, the unsupervised network does not require paired images and it can get a clean image without raindrops. In particular, our network can perform training on paired and unpaired samples. The experimental results show that the best results are achieved not only on the supervised rain-removing network, but also on the unsupervised rain-removing network.

scholarly journals A Multi-Resolution Approach to GAN-Based Speech Enhancement

2021 ◽  
Vol 11 (2) ◽  
pp. 721
Author(s):  
Hyung Yong Kim ◽  
Ji Won Yoon ◽  
Sung Jun Cheon ◽  
Woo Hyun Kang ◽  
Nam Soo Kim
Keyword(s):  
Speech Enhancement ◽  
Optimal Solution ◽  
Experimental Results ◽  
Generative Adversarial Networks ◽  
The Real ◽  
Multi Scale ◽  
Adversarial Networks ◽  
Speech Characteristics ◽  
Conventional Methods ◽  
Convex Property

Recently, generative adversarial networks (GANs) have been successfully applied to speech enhancement. However, there still remain two issues that need to be addressed: (1) GAN-based training is typically unstable due to its non-convex property, and (2) most of the conventional methods do not fully take advantage of the speech characteristics, which could result in a sub-optimal solution. In order to deal with these problems, we propose a progressive generator that can handle the speech in a multi-resolution fashion. Additionally, we propose a multi-scale discriminator that discriminates the real and generated speech at various sampling rates to stabilize GAN training. The proposed structure was compared with the conventional GAN-based speech enhancement algorithms using the VoiceBank-DEMAND dataset. Experimental results showed that the proposed approach can make the training faster and more stable, which improves the performance on various metrics for speech enhancement.

scholarly journals GazeEMD: Detecting Visual Intention in Gaze-Based Human-Robot Interaction

Robotics ◽  
2021 ◽  
Vol 10 (2) ◽  
pp. 68
Author(s):  
Lei Shi ◽  
Cosmin Copot ◽  
Steve Vanlanduit
Keyword(s):  
Cognitive Load ◽  
Real World ◽  
Robotic Manipulator ◽  
Similarity Score ◽  
Human Robot Interaction ◽  
Experimental Results ◽  
The Other ◽  
Robot Interaction ◽  
Run Length ◽  
The Real

In gaze-based Human-Robot Interaction (HRI), it is important to determine human visual intention for interacting with robots. One typical HRI interaction scenario is that a human selects an object by gaze and a robotic manipulator will pick up the object. In this work, we propose an approach, GazeEMD, that can be used to detect whether a human is looking at an object for HRI application. We use Earth Mover’s Distance (EMD) to measure the similarity between the hypothetical gazes at objects and the actual gazes. Then, the similarity score is used to determine if the human visual intention is on the object. We compare our approach with a fixation-based method and HitScan with a run length in the scenario of selecting daily objects by gaze. Our experimental results indicate that the GazeEMD approach has higher accuracy and is more robust to noises than the other approaches. Hence, the users can lessen cognitive load by using our approach in the real-world HRI scenario.

scholarly journals Documentation of the Crime Scene

2015 ◽  
Vol 2 (1) ◽  
pp. 32
Author(s):  
Besim Arifi
Keyword(s):  
Crime Scene ◽  
Final Report ◽  
Real Image ◽  
Crime Scene Investigation ◽  
The Real ◽  
Scene Investigation

Documentation of the scene is one of the most important actions of the investigative group, which performs the examination of the scene. Documentation in itself includes: Taking notes, drawing and photographic team of the site inspection to the crime scene during phase taking notes must pay special attention to this stage, because it includes very important information for the investigation of the crime scene. Based on these information becomes final report on the case, so we can say that taking notes is of particular importance. Another important action is also the sketch. Mapping shows the drawing of the scene and all evidence was found there. Compiler of the sketch doesn’t minds to be definitely professional any painter or designer. It should be a good compiler which presents in detail all the information necessary to include net. This information will be used for testimony in court and to clarify the situation in the scene in case of reconstructions. Photographs of the crime scene investigation is a very important move which reflects the real image of the scene or any part of his and any evidence which has been inserted into focus photography.

scholarly journals Imagology of Iranians in One Thousand Nights and One Night

2016 ◽  
Vol 4 (1) ◽  
pp. 74
Author(s):  
Ehsan Ghabool ◽  
Mina Ravansalar
Keyword(s):  
Comparative Literature ◽  
Literary Work ◽  
Real Image ◽  
Economic Prosperity ◽  
Literary Works ◽  
Arabic Translation ◽  
The Real ◽  
Similarities And Differences ◽  
Definition Of

Imagology is a branch of comparative literature which explores the image of one nation in the literature of another nation. One Thousand Nights and One Night is among the important books which can show the image of different nations and people such as Indians, Iranians and Arabs. Since the oldest version of the book is in Arabic, it is considered an Arabic literary work though it was translated from a Persian tale in the first place. On this basis the study of the image of Iranians in One Thousand Nights and One Night can be included under the definition of imagology. In this article, first we explain, analyze and study the image of Iranians in the book One Thousand Nights and One Night with respect to 1. anthropology (including entertainments, personification of animals, disapprobation of lies and betrayal of spouses), 2. religious and mythical beliefs (including the belief in daevas and jinnis, magic, fire-worshipping and similar plots), 3. politics (emphasizing the position of vizier and his family in government), 4. economics (emphasizing economic prosperity), then we will compare the collected information with the image of Iranians in credited works and in this way we will identify the similarities and differences of Iranians’ image in One Thousand Nights and One Night and the above-said literary works. Finally we come to this conclusion that the similarities belong to the real image of Iranians in the pre-Islamic days and that differences show the image of post-Islamic Iran which is added through Arabic translation.

scholarly journals Multi-Task Self-Supervised Learning for Disfluency Detection

2020 ◽  
Vol 34 (05) ◽  
pp. 9193-9200
Author(s):  
Shaolei Wang ◽  
Wangxiang Che ◽  
Qi Liu ◽  
Pengda Qin ◽  
Ting Liu ◽  
...  
Keyword(s):  
Supervised Learning ◽  
Large Scale ◽  
Experimental Results ◽  
Training Data ◽  
Competitive Performance ◽  
Test Set ◽  
Full Dataset ◽  
Sentence Classification ◽  
Trained Network

Most existing approaches to disfluency detection heavily rely on human-annotated data, which is expensive to obtain in practice. To tackle the training data bottleneck, we investigate methods for combining multiple self-supervised tasks-i.e., supervised tasks where data can be collected without manual labeling. First, we construct large-scale pseudo training data by randomly adding or deleting words from unlabeled news data, and propose two self-supervised pre-training tasks: (i) tagging task to detect the added noisy words. (ii) sentence classification to distinguish original sentences from grammatically-incorrect sentences. We then combine these two tasks to jointly train a network. The pre-trained network is then fine-tuned using human-annotated disfluency detection training data. Experimental results on the commonly used English Switchboard test set show that our approach can achieve competitive performance compared to the previous systems (trained using the full dataset) by using less than 1% (1000 sentences) of the training data. Our method trained on the full dataset significantly outperforms previous methods, reducing the error by 21% on English Switchboard.

scholarly journals Towards Family Economic Sustainability: The Transformation of Traditional Sulphur Miners in Ijen Crater

2021 ◽  
Vol 58 (1) ◽  
pp. 1270-1276
Author(s):  
Alfiana Yuli Efiyanti, Aliya Fatimah
Keyword(s):  
Family Life ◽  
Qualitative Approach ◽  
Economic Sustainability ◽  
Real Image ◽  
Natural Phenomena ◽  
Research Results ◽  
The Real ◽  
Minimum Income ◽  
Permanent Employees

Sulphur miners in Ijen crater, Banyuwangi, Indonesia has been undergoing life changing experience for many years.  It happened due to the natural phenomena of the area.  Mount Ijen has a potential natural tourism that might attract domestic and international tourists.  In addition, the area has high sulphur contents. The management of sulphur is fully authorized by CN Company. The traditional miners deposit the mining results to the company. Working in sulphur mining is very risky, especially for the carriers who are paid based on the quantity of sulphur they bring up from the crater.  This study explores in deep the profile of the traditional sulphur miners in Ijen crater, Banyuwangi. It employs a qualitative approach to obtain the real image of their profile. The research results are (1) three types of traditional sulphur miners in Ijen crater are found. Two of them are permanent employees of the company while the remaining type of the miners is independent worker, whose activities are searching and carrying the sulphur.  The Ijen crater miners  had an uncertain income, depending on the amount of sulphur successfully transported. (2) The minimum income of the traditional miners leads them to transform the job into businesses. (3) The transformation lets them contribute more to their family life.

scholarly journals Concept of “Me” and “Others” in al-Qur’an Perspective

TAWASUT ◽  
2017 ◽  
Vol 1 (1) ◽  
Author(s):  
Muhammad Aniq Imam Muh. Syaifudin
Keyword(s):  
Point Of View ◽  
Real Image ◽  
The Real ◽  
Starting Point

AbstractThe goal of this study is effort to disclose the real image ofislamic teaching particularly its attitude toward others, peaceand dialogue not conflict and is the starting point of mosleminteraction toward others for solving the problems and bringthe point of view close.Keywords: the Qur’an; concept of me and others,tolerance; islamic teaching

Forecasting Groundwater Level Based on Relevance Vector Machine

2010 ◽  
Vol 121-122 ◽  
pp. 43-47 ◽  
Author(s):  
Li Ying Wang ◽  
Wei Guo Zhao
Keyword(s):  
Groundwater Level ◽  
Kernel Method ◽  
Nonlinear Problems ◽  
Relevance Vector Machine ◽  
Kernel Functions ◽  
Experimental Results ◽  
The Real

Relevance Vector Machine (RVM) is a novel kernel method based on sparse Bayesian, which has many advantages such as its kernel functions without the restriction of Mercer’s conditions, and the relevance vectors are automatically determined and have fewer parameters. In this paper, the RVM model is applied to forecasting groundwater level. The experimental results show the final RVM model achieved is sparser, the prediction precision is higher and the prediction values are in better agreement with the real values. It can be concluded that this technique can be seen as a very promising option to solve nonlinear problems such as forecasting groundwater level.

Misrepresentation of Orientals in Dave Eggers’s A Hologram for The King

2021 ◽  
Vol 8 (3) ◽  
pp. 250-258
Author(s):  
Samal Marf Mohammed
Keyword(s):  
Twentieth Century ◽  
Saudi Arabia ◽  
Postcolonial Theory ◽  
Real Image ◽  
Literary Works ◽  
The Real ◽  
Postcolonial Discourse ◽  
Economic Motives ◽  
At Home

      This study deals with the colonial perspectives in Dave Eggers’s A Hologram for The King (2012), according to the postcolonial approach. Although colonialism era is over by now, colonial perspectives remain strong in some literary works. Since its advent in the second half of the twentieth century, postcolonial theory confronts colonial attitudes and experiences as colonialism has been justified in many works of Western writers and scholars who have distorted the real image of non-Europeans and non-Westerners via different means and techniques in masquerade of orientalism. Postcolonial discourse opposes the misrepresentation of non-Europeans and argues that such falsification is driven by political, social, religious and economic motives. In the current study, the researcher aims at explaining the notions of colonialism, otherization and other falsified images of non-Westerners in A Hologram for the King. This paper mainly questions Eggers’s portrayal of the protagonist, Alan Clay, who after bankruptcy and failure at home, flies to Saudi Arabia and capitalizes on the physical and moral assets of the Orientals in this country to convert his story of failure to a success. The characterization of the oriental world and its setting show Eggers’s being biased against the Eastern world and ironically mirror clear hints of colonialism and eurocentrism.

Sign in / Sign up

Export Citation Format

Share Document