检索结果-内蒙古大学图书馆

10th indian conference on computer vision, graphics and image processing (ICVGIP)

作者： Duggal, Rahul Gupta, Anubha Gupta, Ritu Wadhwa, Manya Ahuja, Chirag Indraprastha Inst Informat Technol Delhi Dept ECE SBILab New Delhi India AIIMS Dr BRA IRCH Lab Oncol Unit Delhi India Inst Informat Technol Delhi Dept CSE New Delhi India

ISBN: (纸本)9781450347532

this paper proposes a method for segmentation of nuclei of single/isolated and overlapping/touching immature white blood cells from microscopic images of B-Lineage acute lymphoblastic leukemia (ALL) prepared from peripheral blood and bone marrow aspirate. We propose deep belief network approach for the segmentation of these nuclei. Simulation results and comparison with some of the existing methods demonstrate the efficacy of the proposed method.

关键词： Overlapping nuclei segmentation B-ALL Deep Belief Network Machine Learning

来源：评论

学校读者我要写书评

暂无评论

Lip Contour Detection for Estimation of Mouth Opening Area 5

Lip Contour Detection for Estimation of Mouth Opening Area

引用

5th National conference on computer vision, Pattern Recognition, image processing and graphics (NCVPRIPG)

作者： Jain, Shreyansh Pandey, Prem C. Velmurugan, Rajbabu Indian Inst Technol Dept Elect Engn Bombay 400076 Maharashtra India

ISBN: (纸本)9781467385640

In speech training aids for providing visual feedback of the articulatory efforts, time-varying vocal tract shape during speech production is generally obtained by linear prediction (LP) analysis of the speech signal and assuming a constant area at the glottis end as a reference. Its variation during speech production causes errors in the estimated vocal tract shape. the problem can be overcome by using area of the mouth opening as the reference. this area can be estimated by detecting the inner lip contour from the video recording of speaker's face during speech utterance. A technique for detection of inner lip contour, based on color transformation and template matching, is presented for reducing the errors caused by presence of teeth and tongue. Face detection by Viola-Jones algorithm, localization using a mouth detection technique, and outer lip contour detection are used to narrow down the search region for inner mouth opening. Presence of the teeth is masked by separate color transformations for upper and lower lip segments. For reducing the errors due to visibility of the tongue, which may not have any significant separation from the lips in the color space, a template matching technique is employed. It is used separately for the upper and lower lip segments to obtain the mouth opening area. the technique has been validated against graphically measured values of the mouth opening and found to be successful in estimating the mouth opening area, and it is not affected by skin hue and presence of teeth.

关键词： lip contour mouth opening area speech training aid template matching

来源：评论

学校读者我要写书评

暂无评论

Feature-Preserving 3D Fluorescence image Sequence Denoising 16

Feature-Preserving 3D Fluorescence Image Sequence Denoising

引用

10th indian conference on computer vision, graphics and image processing (ICVGIP)

作者： Bhujle, Hemalata SDM Coll Engn & Technol Dept Elect & Commun Engn Dharwad Karnataka India

ISBN: (纸本)9781450347532

In this paper feature-preserving denoising scheme for fluorescence video microscopy is presented. Fluorescence image sequences comprise of edges and fine structures with fast moving objects. Improving signal to noise ratio (SNR) while preserving structural details is a difficult task for these image sequences. Few existing denoising techniques result in over smoothing these image sequences while others fail due to inappropriate implementation of motion estimation and compensation steps. In this paper we use nonlocal means (NLM) video denoising algorithm as to avoid motion estimation and compensation steps. the proposed shot boundary detection technique pre-processes the sequence systematically and accurately to form different shots with content-wise similar frames. To preserve the edges and fine structural details in the image sequences we modify the weighing term of NLM filter. Further, to accelerate the denoising process, separable non-local means filter is implemented for video sequences. We compare the results with existing fluorescence video denoising techniques and show that the proposed method not only preserves the edges and small structural details more efficiently, also reduces the computational time. Efficacy of the proposed algorithm is evaluated quantitatively and qualitatively with PSNR and vision perception.

关键词： Non-local means fluorescence microscopy

来源：评论

学校读者我要写书评

暂无评论

Towards Semantic Visual Representation: Augmenting image Representation with Natural Language Descriptors 16

Towards Semantic Visual Representation: Augmenting Image Rep...

引用

10th indian conference on computer vision, graphics and image processing (ICVGIP)

作者： Mopuri, Konda Reddy Babu, R. Venkatesh Indian Inst Sci Bangalore Karnataka India

ISBN: (纸本)9781450347532

Learning image representations has been an interesting and challenging problem. When users upload images to photo sharing websites, they often provide multiple textual tags for ease of reference. these tags can reveal significant information about the content of the image such as the objects present in the image or the action that is taking place. Approaches have been proposed to extract additional information from these tags in order to augment the visual cues and build a multi-modal image representation. However, the existing approaches do not pay much attention to the semantic meaning of the tags while they encode. In this work, we attempt to enrich the image representation with the tag encodings that leverage their semantics. Our approach utilizes neural network based natural language descriptors to represent the tag information. By complementing the visual features learned by convnets, our approach results in an efficient multi-modal image representation. Experimental evaluation suggests that our approach results in a better multi-modal image representation by exploiting the two data modalities for classification on benchmark datasets.

关键词： Multi-modal image representation neural embeddings natural language descriptors word2vec text encoding image description

来源：评论

学校读者我要写书评

暂无评论

Event Recognition in Broadcast Soccer Videos 16

Event Recognition in Broadcast Soccer Videos

引用

10th indian conference on computer vision, graphics and image processing (ICVGIP)

作者： Saraogi, Himangi Sharma, Rahul Anand Kumar, Vijay IIIT Hyderabad Ctr Visual Informat Technol Hyderabad Telangana India

ISBN: (纸本)9781450347532

Automatic recognition of important events in soccer broadcast videos plays a vital role in many applications including video summarization, indexing, content-based search, and in performance analysis of players and teams. this paper proposes an approach for soccer event recognition using deep convolutional features combined with domain-specific cues. For deep representation, we use the recently proposed trajectory based deep convolutional descriptor (TDD) [1] which samples and pools the discriminatively trained convolutional features around the improved trajectories. We further improve the performance by incorporating domain specific knowledge based on camera view type and its position. the camera position and view type captures the statistics of occurrence of events in different play-field regions and zoom-level respectively. We conduct extensive experiments on 6 hour long soccer matches and show the effectiveness of deep video representation for soccer and the improvements obtained using domain-specific cues.

关键词： Soccer event recognition deep convolutional features playground registeration view-shot estimation

来源：评论

学校读者我要写书评

暂无评论

On the view synthesis of man-made scenes using uncalibrated cameras

引用

PATTERN RECOGNITION LETTERS 2007年第3期28卷 384-393页

作者： Sharma, Geetika Chaudhury, Santanu Srivastava, J. B. Indian Inst Technol Dept Elect Engn New Delhi 110016 India Indian Inst Technol Dept Math New Delhi 110016 India

We have attempted the problem of novel view synthesis of scenes containing man-made objects from images taken by arbitrary, uncalibrated cameras. Under the assumption of availability of the correspondence of three vanishing points, in general position, we propose two techniques. the first is a transfer-based scheme which synthesizes new views with only a translation of the virtual camera and computes z-buffer values for handling occlusions in synthesized views. the second is a reconstruction-based scheme which synthesizes arbitrary new views in which the camera can undergo rotation as well as translation. We present experimental results to establish the validity of both formulations. (c) 2006 Published by Elsevier B.V.

关键词： image-based rendering novel view synthesis uncalibrated cameras

来源：评论

学校读者我要写书评

暂无评论

Adaptive Artistic Stylization of images 16

Adaptive Artistic Stylization of Images

引用

10th indian conference on computer vision, graphics and image processing (ICVGIP)

作者： Deshpande, Ameya Raman, Shanmuganathan IIT Gandhinagar Gandhinagar Gujarat India

ISBN: (纸本)9781450347532

In this work, we present a novel non-photorealistic rendering method which produces good quality stylization results for color images. the procedure is driven by saliency measure in the foreground and the background region. We start with generating saliency map and simple thresholding based segmentation to get rough estimation of the foreground background mask. We improve this mask by using a scribble based method where the scribbles for foreground-background regions are automatically generated from the previous rough estimation. Followed by the mask generation, we proceed with an iterative abstraction process which involves edge preserving blurring and edge detection. the number of iterations of the abstraction process to be performed in the foreground and background regions are decided by tracking the changes in saliency measure in the foreground and the background regions. Performing unequal number of iterations helps to improve the average saliency measure in more salient region (foreground) while decreasing the average saliency measure in the non-salient region (background). Implementation results of our method shows the merits of this approach with other competing methods.

关键词： Non-photorealistic rendering Saliency image Abstraction Guided Filter

来源：评论

学校读者我要写书评

暂无评论

SpoofNET: Resolving Facial Makeup based Spoofs 11

SpoofNET: Resolving Facial Makeup based Spoofs

引用

11th indian conference on computer vision, graphics and image processing (ICVGIP)

作者： Bhattacharjee, Avishek Banerjee, Samik Das, Sukhendu Indian Inst Technol Madras Chennai 600036 Tamil Nadu India

ISBN: (纸本)9781450366151

Face Recognition (FR) under adversarial conditions has been a big challenge for researchers in the computer vision and Machine Learning communities in the recent past. Most of state-of-the-art face recognition systems have been designed to overcome degradations in a face due to variations in pose, illumination, contrast, resolution, along with blur. However, interestingly none have addressed the fascinating issue of makeup as a spoof attack, which drastically changes the appearance of a face, making it difficult for even humans to detect and identify the impostor. In this paper, we propose a novel multi-component deep convolutional neural network (CNN) based architecture which performs the complex task of makeup removal from a disguised face, to reveal the original mugshot image of the impostor (i.e. without makeup). the proposed network also performs the hard tasks of FR on a disguised face in addition to recognition of identity and generation of the face of the spoofed target, by minimizing a novel multi-component objective function. Comparison of performance with a few recent state-of-the-art methods of FR over three benchmark datasets reveals the superiority of our proposed method for both synthesis as well as recognition (FR) tasks.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Multimodal Egocentric Activity Recognition Using Multi-stream CNN 11

Multimodal Egocentric Activity Recognition Using Multi-strea...

引用

11th indian conference on computer vision, graphics and image processing (ICVGIP)

作者： Imran, Javed Raman, Balasubramanian Indian Inst Technol Roorkee Dept Comp Sci & Engn Roorkee Uttar Pradesh India

ISBN: (纸本)9781450366151

Egocentric activity recognition (EAR) is an emerging area in the field of computer vision research. Motivated by the current success of Convolutional Neural Network (CNN), we propose a multi-stream CNN for multimodal egocentric activity recognition using visual (RGB videos) and sensor stream (accelerometer, gyroscope, etc.). In order to effectively capture the spatio-temporal information contained in RGB videos, two types of modalities are extracted from visual data: Approximate Dynamic image (ADI) and Stacked Difference image (SDI). these image-based representations are generated both at clip level as well as entire video level, and are then utilized to finetune a pretrained 2D-CNN called MobileNet, which is specifically designed for mobile vision applications. Similarly for sensor data, each training sample is divided into three segments, and a deep 1D-CNN network is trained (corresponding to each type of sensor stream) from scratch. During testing, the softmax scores of all the streams (visual + sensor) are combined by late fusion. the experiments performed on multimodal egocentric activity dataset demonstrates that our proposed approach can achieve state-of-the-art results, outperforming the current best handcrafted and deep learning based techniques.

关键词： Egocentric Activity Recognition Convolutional Neural Network Dynamic image Stacked Difference image Multimodal Fusion

来源：评论

学校读者我要写书评

暂无评论

A Context Dependent Fragile Watermarking Scheme for Tamper Detection from Demosaicked Color images 16

A Context Dependent Fragile Watermarking Scheme for Tamper D...

引用

10th indian conference on computer vision, graphics and image processing (ICVGIP)

作者： Manikandan, V. M. Masilamani, V. Indian Inst Informat Technol Design & Mfg Kanchee Madras 600127 Tamil Nadu India

ISBN: (纸本)9781450347532

Fragile watermarking schemes are very common in practice for tamper detection and image authentication. In this paper, we propose a new fragile watermarking scheme to detect common image tampering operations such as copy and paste attack and image splicing attack from color images. the proposed scheme is intended to do the watermarking with an image dependent authentication code during color image demosaicking procedure. We propose a new method to generate almost unique authentication code from an image by comparing RGB color components of the pixels. In the proposed scheme, every pixel will be watermarked with an encrypted authentication code derived from the same pixel. To embed the watermark a new method has been proposed in this paper, where the color filter array sampled components will not be considered during watermarking, and only the rebuilt color component values of every pixel will be modified according to the authentication code. the experimental study shows that the proposed watermarking scheme produces images with better visual quality as compared to the state-of-the-art, and is capable of detecting both copy and paste attacks and image splicing attacks more accurately than the existing schemes.

关键词： fragile watermarking image demosaicking tamper detection authentication code

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：