Search Results: 1 - 10 of 100 matches for " "
All listed articles are free for downloading (OA Articles)
Page 1 /100
Display every page Item
Temporal Localization of Fine-Grained Actions in Videos by Domain Transfer from Web Images  [PDF]
Chen Sun,Sanketh Shetty,Rahul Sukthankar,Ram Nevatia
Computer Science , 2015, DOI: 10.1145/2733373.2806226
Abstract: We address the problem of fine-grained action localization from temporally untrimmed web videos. We assume that only weak video-level annotations are available for training. The goal is to use these weak labels to identify temporal segments corresponding to the actions, and learn models that generalize to unconstrained web videos. We find that web images queried by action names serve as well-localized highlights for many actions, but are noisily labeled. To solve this problem, we propose a simple yet effective method that takes weak video labels and noisy image labels as input, and generates localized action frames as output. This is achieved by cross-domain transfer between video frames and web images, using pre-trained deep convolutional neural networks. We then use the localized action frames to train action recognition models with long short-term memory networks. We collect a fine-grained sports action data set FGA-240 of more than 130,000 YouTube videos. It has 240 fine-grained actions under 85 sports activities. Convincing results are shown on the FGA-240 data set, as well as the THUMOS 2014 localization data set with untrimmed training videos.
Anti-compression approach to distinguishing photographic images and photorealistic computer graphics

Zhang Rong,Wang Rangding,

中国图象图形学报 , 2012,
Abstract: Based on the sparse representation of images, a new approach to distinguish photographic images and photorealistic computer graphics is proposed. The proposed approach is robust enough to compression to guarantee image authenticity forensics. The Tetrolet transformation chooses the optimal tetromino partition for each 4×4 image block in terms of the minimal L1-norm criterion to protect local image geometry structure and to obtain the sparsest image representation. When observing the adaptive values c, an image is represented as a normalized histogram with 117 bins corresponding to the number of occurrences of different block covering, i.e. the features of HoC (Histogram of Covering). The experimental results demonstrate the HoC features extracted from S (saturation) are able to characterize the distinct statistical properties in the local geometry between photographic images and photorealistic computer graphics. The proposed approach is applicable to image authenticity detection and auto-classification.
Automatic Color Transfer of Images  [PDF]
Guoliang Mo,Sanyuan Zhang,Xiaocong Zhou,Jialin Yan
Information Technology Journal , 2012,
Abstract: Color transfer of images means transferring color characteristics from one image to another, so that the target image and source image have similar colors. This study proposed a method of automatic color transfer of sampling images. The main steps are as follows. First, based on the unsupervised clustering, divide the target images and source images into a number of subsets with similar color and extract sample points with higher density from each subset to form sample sets. Then according to the definition of similarity measure, the correspondence between the sample sets is established. Third, use the best matching algorithm to find the best matching samples of the target sample set in the source sample set and pass the color information to the target sample points. At last, take color transfer of the non-sampled part in the target image with the sampled set which has been taken color transfer. Experimental results show that, although there is no interactive sampling operation, this method can approximate the results of the color transfer with interactive operation.
Photorealistic Modeling of the Growth of Filamentous Specimens  [cached]
Ji?í Sedlá?,Jan Flusser,Michaela Sedlá?ová
EURASIP Journal on Advances in Signal Processing , 2008, DOI: 10.1155/2008/520972
Abstract: We present a new method for modeling the development of settled specimens with filamentous growth patterns, such as fungi and oomycetes. In phytopathology, the growth parameters of such microorganisms are frequently examined. Their development is documented repeatedly, in a defined time sequence, leaving the growth pattern incomplete. This restriction can be overcome by reconstructing the missing images from the images acquired at consecutive observation sessions. Image warping is a convenient tool for such purposes. In the proposed method, the parameters of the geometric transformation are estimated by means of the growth tracking based on the morphological skeleton. The result is a sequence of photorealistic artificial images that show the development of the specimen within the interval between observations.
Fine properties of Charge Transfer Models  [PDF]
Kaihua Cai
Mathematics , 2003,
Abstract: We prove L^1 to L^infinity estimates for charge transfer Hamiltonians H(t) in R^n for n > or = 3, followed by a discussion of estimates from W^{k,p'} to W^{k,p} for the same model, where 2 < p < infinity and 1/p + 1/p'=1. Then, geometric methods are developed to establish the time boundedness of the H^k norm for the evolution of charge transfer operators and asymptotic completeness of the Hamiltonian H(t) in the H^k norm, where k is any positive integer.
Computer-Generated Photorealistic Hair  [PDF]
Alice J. Lin
Computer Science , 2002,
Abstract: This paper presents an efficient method for generating and rendering photorealistic hair in two dimensional pictures. The method consists of three major steps. Simulating an artist drawing is used to design the rough hair shape. A convolution based filter is then used to generate photorealistic hair patches. A refine procedure is finally used to blend the boundaries of the patches with surrounding areas. This method can be used to create all types of photorealistic human hair (head hair, facial hair and body hair). It is also suitable for fur and grass generation. Applications of this method include: hairstyle designing/editing, damaged hair image restoration, human hair animation, virtual makeover of a human, and landscape creation.
Fine mapping of the awn gene on chromosome 4 in rice by association and linkage analyses
GuangLong Hu,DongLing Zhang,HuiQiao Pan,Ben Li,JianTao Wu,XueYa Zhou,QunYuan Zhang,Lei Zhou,GuoXin Yao,JunZhou Li,JinJie Li,HongLiang Zhang,ZiChao Li
Chinese Science Bulletin , 2011, DOI: 10.1007/s11434-010-4181-5
Abstract: Awnness is a key trait in rice domestication, yet no studies have been conducted on fine mapping or association mapping of the rice awn gene. In this study, we investigated the awnness and genotype of a core collection of 303 cultivated rice varieties and a BC5F2 segregating population of 200 individuals. Combining association and linkage analyses, we mapped the awnness related genes to chromosome 4. Primary association analysis using 24 SSR markers revealed five loci significantly associated with awnness on chromosome 4. The associated markers cover previously identified regions. Fine association mapping was conducted using another 29 markers within a 4-Mb region, covering the associated marker in34, which is close to the awn gene Awn4.1. Seven associated markers were revealed, distributed over an 870-kb region. Combining the fine association mapping and linkage analysis of awnness in the 200 BC5F2 segregating population, we finally identified a 330-kb region as the candidate region for Awn4.1. The results indicate that combining association mapping and linkage mapping provides an efficient and precise approach to both genome-wide mapping and fine mapping of rice genes.
Variation of the fine-structure constant: an update of statistical analyses with recent data  [PDF]
Lucila Kraiselburd,Susana J. Landau,Claudio Simeone
Physics , 2013, DOI: 10.1051/0004-6361/201321803
Abstract: We analyze the consistency of different astronomical data of the variation in the fine-structure constant obtained with KECK and VLT. We tested the consistency using the Student test and confidence intervals. We splited the data sets in to smaller intervals and grouped them acording to redshift and angular position. Another statistical analysis is proposed that considers phenomenological models for the variation in \alpha. Results show consistency for the reduced intervals for each pair of data sets and suggests that the variation in \alpha is significant at higher redshifts. Even though the dipole model seems to be the most accurate phenomenological model, the statistical analyses indicate that the variation in \alpha might be depending on both redshift and angular position.
Efficient Upsampling of Natural Images  [PDF]
Chinmay Hegde,Oncel Tuzel,Fatih Porikli
Computer Science , 2015,
Abstract: We propose a novel method of efficient upsampling of a single natural image. Current methods for image upsampling tend to produce high-resolution images with either blurry salient edges, or loss of fine textural detail, or spurious noise artifacts. In our method, we mitigate these effects by modeling the input image as a sum of edge and detail layers, operating upon these layers separately, and merging the upscaled results in an automatic fashion. We formulate the upsampled output image as the solution to a non-convex energy minimization problem, and propose an algorithm to obtain a tractable approximate solution. Our algorithm comprises two main stages. 1) For the edge layer, we use a nonparametric approach by constructing a dictionary of patches from a given image, and synthesize edge regions in a higher-resolution version of the image. 2) For the detail layer, we use a global parametric texture enhancement approach to synthesize detail regions across the image. We demonstrate that our method is able to accurately reproduce sharp edges as well as synthesize photorealistic textures, while avoiding common artifacts such as ringing and haloing. In addition, our method involves no training phase or estimation of model parameters, and is easily parallelizable. We demonstrate the utility of our method on a number of challenging standard test photos.
Neoliberalism in education: Five images of critical analyses  [PDF]
Ondrej Ka??ák,Branislav Pupala
Pedagogicka Orientace , 2011,
Abstract: The survey study brings information about the way that educational researchcopes with neoliberalism as a generalized form of social government in the currentwestern culture. It shows that neoliberalism is considered as a universal scope of otherchanges in the basic segments of education and those theoretical and critical analyses ofthis phenomenon represent an important part of production in the area of educationalresearch. It emphasizes the contribution of formation and development of the socalledgovernmental studies for comprehension of mechanisms and consequences ofneoliberal government of the society and shows how way the methodology of thesestudies helps to identify neoliberal strategies used in the regulation of social subjectsby education. There are five selected segments of critical analyses elaborated (fromthe concept of a lifelong learning, through preschool and university education to theeducation of teachers and PISA project) that obviously show ideological and theoreticalcohesiveness of the education analysis through the scope of neoliberal governmentality.
Page 1 /100
Display every page Item

Copyright © 2008-2017 Open Access Library. All rights reserved.