Sorry, you need to enable JavaScript to visit this website.

Image/Video Processing

How Should We Handle 4D Light Fields with CNNs?


We investigated how we should handle high dimensional light fields (LFs) with convolutional neural networks (CNNs).
An LF is a 4-D signal representation of light rays, and it is interpreted as a set of dense multi-view images.
As an important building block of various light field applications, we focused on signal restoration problems for LFs, and we adopted CNNs as the solver for them because of its striking performance on the conventional 2-D images.
In applying CNNs, the high dimensionality of LFs should be carefully addressed.

Paper Details

Authors:
Shu Fujita, Keita Takahashi, Toshiaki Fujii
Submitted On:
5 October 2018 - 12:39am
Short Link:
Type:

Document Files

icip_fujita_poster.pdf

(27 downloads)

Subscribe

[1] Shu Fujita, Keita Takahashi, Toshiaki Fujii, "How Should We Handle 4D Light Fields with CNNs?", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3491. Accessed: Jan. 16, 2019.
@article{3491-18,
url = {http://sigport.org/3491},
author = {Shu Fujita; Keita Takahashi; Toshiaki Fujii },
publisher = {IEEE SigPort},
title = {How Should We Handle 4D Light Fields with CNNs?},
year = {2018} }
TY - EJOUR
T1 - How Should We Handle 4D Light Fields with CNNs?
AU - Shu Fujita; Keita Takahashi; Toshiaki Fujii
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3491
ER -
Shu Fujita, Keita Takahashi, Toshiaki Fujii. (2018). How Should We Handle 4D Light Fields with CNNs?. IEEE SigPort. http://sigport.org/3491
Shu Fujita, Keita Takahashi, Toshiaki Fujii, 2018. How Should We Handle 4D Light Fields with CNNs?. Available at: http://sigport.org/3491.
Shu Fujita, Keita Takahashi, Toshiaki Fujii. (2018). "How Should We Handle 4D Light Fields with CNNs?." Web.
1. Shu Fujita, Keita Takahashi, Toshiaki Fujii. How Should We Handle 4D Light Fields with CNNs? [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3491

CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION


Image processing has been a popular tool for biological researches. Detecting specific animals in aerial images captured by an UAV is a crucial research topic. As the rapid progress of deep learning (DL), it has been a popular approach to many image classification and object detection tasks. However, DL usually requires a large set of training samples to learn the network weights, while the biological image materials are often insufficient to fulfill the demand.

Paper Details

Authors:
Yi-Min Chou, Chien-Hung Chen, Keng-Hao Liu, Chu-Song Chen
Submitted On:
5 October 2018 - 12:00am
Short Link:
Type:
Event:
Document Year:
Cite

Document Files

CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION

(24 downloads)

Subscribe

[1] Yi-Min Chou, Chien-Hung Chen, Keng-Hao Liu, Chu-Song Chen, "CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3486. Accessed: Jan. 16, 2019.
@article{3486-18,
url = {http://sigport.org/3486},
author = {Yi-Min Chou; Chien-Hung Chen; Keng-Hao Liu; Chu-Song Chen },
publisher = {IEEE SigPort},
title = {CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION},
year = {2018} }
TY - EJOUR
T1 - CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION
AU - Yi-Min Chou; Chien-Hung Chen; Keng-Hao Liu; Chu-Song Chen
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3486
ER -
Yi-Min Chou, Chien-Hung Chen, Keng-Hao Liu, Chu-Song Chen. (2018). CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION. IEEE SigPort. http://sigport.org/3486
Yi-Min Chou, Chien-Hung Chen, Keng-Hao Liu, Chu-Song Chen, 2018. CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION. Available at: http://sigport.org/3486.
Yi-Min Chou, Chien-Hung Chen, Keng-Hao Liu, Chu-Song Chen. (2018). "CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION." Web.
1. Yi-Min Chou, Chien-Hung Chen, Keng-Hao Liu, Chu-Song Chen. CHANGING BACKGROUND TO FOREGROUND: AN AUGMENTATION METHOD BASED ON CONDITIONAL GENERATIVE NETWORK FOR STINGRAY DETECTION [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3486

Double Complete D-LBP with Extreme Learning Machine Auto-Encoder and Cascade Forest for Facial Expression Analysis


Although the obtained accuracy on some lab-controlled facial expression datasets has been very high, the recognition of facial expressions in wild environments is still a challenging problem. Local Binary Patterns (LBP) is a widely used operator in facial expression recognition. However, there are few variations of LBP operators specifically designed for facial expression recognition. In this paper, we propose a novel representation approach called the Double Complete d-LBP (Double Cd-LBP) according to the characteristics of facial expressions.

Paper Details

Authors:
Jing Liu,Peng Wu
Submitted On:
4 October 2018 - 11:23pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2018-1984-poster.pdf

(28 downloads)

Subscribe

[1] Jing Liu,Peng Wu, "Double Complete D-LBP with Extreme Learning Machine Auto-Encoder and Cascade Forest for Facial Expression Analysis", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3485. Accessed: Jan. 16, 2019.
@article{3485-18,
url = {http://sigport.org/3485},
author = {Jing Liu;Peng Wu },
publisher = {IEEE SigPort},
title = {Double Complete D-LBP with Extreme Learning Machine Auto-Encoder and Cascade Forest for Facial Expression Analysis},
year = {2018} }
TY - EJOUR
T1 - Double Complete D-LBP with Extreme Learning Machine Auto-Encoder and Cascade Forest for Facial Expression Analysis
AU - Jing Liu;Peng Wu
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3485
ER -
Jing Liu,Peng Wu. (2018). Double Complete D-LBP with Extreme Learning Machine Auto-Encoder and Cascade Forest for Facial Expression Analysis. IEEE SigPort. http://sigport.org/3485
Jing Liu,Peng Wu, 2018. Double Complete D-LBP with Extreme Learning Machine Auto-Encoder and Cascade Forest for Facial Expression Analysis. Available at: http://sigport.org/3485.
Jing Liu,Peng Wu. (2018). "Double Complete D-LBP with Extreme Learning Machine Auto-Encoder and Cascade Forest for Facial Expression Analysis." Web.
1. Jing Liu,Peng Wu. Double Complete D-LBP with Extreme Learning Machine Auto-Encoder and Cascade Forest for Facial Expression Analysis [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3485

Automatic Generation of Epipolar Curves


Fisheye camera is widely used in various applications because of its wider field-of-view. However, high distortion of images captured by fisheye cameras make it difficult for certain tasks which are based on traditional epipolar geometry (using epipolar lines) and stereo correspondence, such as depth map estimation. While most of existing depth map estimation methods use perspective-projection-based camera model, considering fisheye camera for depth map estimation will be beneficial because of its wider FOV.

Paper Details

Authors:
Yi-Yu Hsieh, Jiwa Malem Marsya, Jen-Hui Chuang
Submitted On:
4 October 2018 - 11:20pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2018_EpipolarCurves_Poster_Final.pdf

(32 downloads)

Subscribe

[1] Yi-Yu Hsieh, Jiwa Malem Marsya, Jen-Hui Chuang, "Automatic Generation of Epipolar Curves", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3484. Accessed: Jan. 16, 2019.
@article{3484-18,
url = {http://sigport.org/3484},
author = {Yi-Yu Hsieh; Jiwa Malem Marsya; Jen-Hui Chuang },
publisher = {IEEE SigPort},
title = {Automatic Generation of Epipolar Curves},
year = {2018} }
TY - EJOUR
T1 - Automatic Generation of Epipolar Curves
AU - Yi-Yu Hsieh; Jiwa Malem Marsya; Jen-Hui Chuang
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3484
ER -
Yi-Yu Hsieh, Jiwa Malem Marsya, Jen-Hui Chuang. (2018). Automatic Generation of Epipolar Curves. IEEE SigPort. http://sigport.org/3484
Yi-Yu Hsieh, Jiwa Malem Marsya, Jen-Hui Chuang, 2018. Automatic Generation of Epipolar Curves. Available at: http://sigport.org/3484.
Yi-Yu Hsieh, Jiwa Malem Marsya, Jen-Hui Chuang. (2018). "Automatic Generation of Epipolar Curves." Web.
1. Yi-Yu Hsieh, Jiwa Malem Marsya, Jen-Hui Chuang. Automatic Generation of Epipolar Curves [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3484

Image Sharpness Metric Based on MaxPol Convolution Kernels


We presents a no-reference (NR) image sharpness metric based on a visual sensitivity model. We propose that MaxPol convolution kernels are close approximation to this model and capable of extracting meaningful features for image sharpness assessment. Equipped by these kernels, we develop an efficient pipeline to evaluate the out-of-focus level of input images by decomposing the first and third order image differentials. The associated kernels are regulated in higher cutoff frequencies to balance out the information loss and noise sensitivity.

Paper Details

Authors:
Mahdi S. Hosseini and Konstantinos N. Plataniotis
Submitted On:
4 October 2018 - 11:06pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Presentation Slides for ICIP2018

(45 downloads)

Subscribe

[1] Mahdi S. Hosseini and Konstantinos N. Plataniotis, "Image Sharpness Metric Based on MaxPol Convolution Kernels", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3482. Accessed: Jan. 16, 2019.
@article{3482-18,
url = {http://sigport.org/3482},
author = {Mahdi S. Hosseini and Konstantinos N. Plataniotis },
publisher = {IEEE SigPort},
title = {Image Sharpness Metric Based on MaxPol Convolution Kernels},
year = {2018} }
TY - EJOUR
T1 - Image Sharpness Metric Based on MaxPol Convolution Kernels
AU - Mahdi S. Hosseini and Konstantinos N. Plataniotis
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3482
ER -
Mahdi S. Hosseini and Konstantinos N. Plataniotis. (2018). Image Sharpness Metric Based on MaxPol Convolution Kernels. IEEE SigPort. http://sigport.org/3482
Mahdi S. Hosseini and Konstantinos N. Plataniotis, 2018. Image Sharpness Metric Based on MaxPol Convolution Kernels. Available at: http://sigport.org/3482.
Mahdi S. Hosseini and Konstantinos N. Plataniotis. (2018). "Image Sharpness Metric Based on MaxPol Convolution Kernels." Web.
1. Mahdi S. Hosseini and Konstantinos N. Plataniotis. Image Sharpness Metric Based on MaxPol Convolution Kernels [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3482

HYPERSPECTRAL IMAGERY DENOISING VIA REWEIGHED SPARSE LOW-RANK NONNEGATIVE TENSOR FACTORIZATION

Paper Details

Authors:
Fengchao Xiong, Jun Zhou
Submitted On:
4 October 2018 - 11:00pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2018-2782-poster.pdf

(19 downloads)

Subscribe

[1] Fengchao Xiong, Jun Zhou, "HYPERSPECTRAL IMAGERY DENOISING VIA REWEIGHED SPARSE LOW-RANK NONNEGATIVE TENSOR FACTORIZATION", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3481. Accessed: Jan. 16, 2019.
@article{3481-18,
url = {http://sigport.org/3481},
author = {Fengchao Xiong; Jun Zhou },
publisher = {IEEE SigPort},
title = {HYPERSPECTRAL IMAGERY DENOISING VIA REWEIGHED SPARSE LOW-RANK NONNEGATIVE TENSOR FACTORIZATION},
year = {2018} }
TY - EJOUR
T1 - HYPERSPECTRAL IMAGERY DENOISING VIA REWEIGHED SPARSE LOW-RANK NONNEGATIVE TENSOR FACTORIZATION
AU - Fengchao Xiong; Jun Zhou
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3481
ER -
Fengchao Xiong, Jun Zhou. (2018). HYPERSPECTRAL IMAGERY DENOISING VIA REWEIGHED SPARSE LOW-RANK NONNEGATIVE TENSOR FACTORIZATION. IEEE SigPort. http://sigport.org/3481
Fengchao Xiong, Jun Zhou, 2018. HYPERSPECTRAL IMAGERY DENOISING VIA REWEIGHED SPARSE LOW-RANK NONNEGATIVE TENSOR FACTORIZATION. Available at: http://sigport.org/3481.
Fengchao Xiong, Jun Zhou. (2018). "HYPERSPECTRAL IMAGERY DENOISING VIA REWEIGHED SPARSE LOW-RANK NONNEGATIVE TENSOR FACTORIZATION." Web.
1. Fengchao Xiong, Jun Zhou. HYPERSPECTRAL IMAGERY DENOISING VIA REWEIGHED SPARSE LOW-RANK NONNEGATIVE TENSOR FACTORIZATION [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3481

FAST AND ROBUST VANISHING POINT DETECTION ON UN-CALIBRATED IMAGES


This paper presents a novel algorithm for fast and effective vanishing point detection. Once line segments in an input image are detected by LSD algorithm, the proposed method filters out outlier line segments. The remaining line segments are then over-clustered, and each cluster is assigned to 5 different types. According to the assigned type, each cluster is re-merged by applying different criteria, and the re-merged clusters generate hypotheses for vanishing points.

Paper Details

Authors:
Sang Jun Lee, Sung Soo Hwang
Submitted On:
4 October 2018 - 10:33pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2018Poster.pdf

(65 downloads)

Subscribe

[1] Sang Jun Lee, Sung Soo Hwang, "FAST AND ROBUST VANISHING POINT DETECTION ON UN-CALIBRATED IMAGES", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3480. Accessed: Jan. 16, 2019.
@article{3480-18,
url = {http://sigport.org/3480},
author = {Sang Jun Lee; Sung Soo Hwang },
publisher = {IEEE SigPort},
title = {FAST AND ROBUST VANISHING POINT DETECTION ON UN-CALIBRATED IMAGES},
year = {2018} }
TY - EJOUR
T1 - FAST AND ROBUST VANISHING POINT DETECTION ON UN-CALIBRATED IMAGES
AU - Sang Jun Lee; Sung Soo Hwang
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3480
ER -
Sang Jun Lee, Sung Soo Hwang. (2018). FAST AND ROBUST VANISHING POINT DETECTION ON UN-CALIBRATED IMAGES. IEEE SigPort. http://sigport.org/3480
Sang Jun Lee, Sung Soo Hwang, 2018. FAST AND ROBUST VANISHING POINT DETECTION ON UN-CALIBRATED IMAGES. Available at: http://sigport.org/3480.
Sang Jun Lee, Sung Soo Hwang. (2018). "FAST AND ROBUST VANISHING POINT DETECTION ON UN-CALIBRATED IMAGES." Web.
1. Sang Jun Lee, Sung Soo Hwang. FAST AND ROBUST VANISHING POINT DETECTION ON UN-CALIBRATED IMAGES [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3480

Closed-form Solution of Simultaneous Denoising and Hole Filling of Depth Image


This paper presents a novel Time-of-Flight (ToF) depth recovery algorithm minimizing a new quadratic energy function utilizing depth and infrared data. The proposed energy function consists of a filtering term and a reconstruction term to remove noise and fill holes simultaneously in a depth image. In the filtering term, a new multilateral weight is introduced by fully using available spatial, depth, and infra¬red information.

Paper Details

Authors:
Submitted On:
4 October 2018 - 9:21pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

icip2018_yskim.pdf

(18 downloads)

Subscribe

[1] , "Closed-form Solution of Simultaneous Denoising and Hole Filling of Depth Image", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3473. Accessed: Jan. 16, 2019.
@article{3473-18,
url = {http://sigport.org/3473},
author = { },
publisher = {IEEE SigPort},
title = {Closed-form Solution of Simultaneous Denoising and Hole Filling of Depth Image},
year = {2018} }
TY - EJOUR
T1 - Closed-form Solution of Simultaneous Denoising and Hole Filling of Depth Image
AU -
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3473
ER -
. (2018). Closed-form Solution of Simultaneous Denoising and Hole Filling of Depth Image. IEEE SigPort. http://sigport.org/3473
, 2018. Closed-form Solution of Simultaneous Denoising and Hole Filling of Depth Image. Available at: http://sigport.org/3473.
. (2018). "Closed-form Solution of Simultaneous Denoising and Hole Filling of Depth Image." Web.
1. . Closed-form Solution of Simultaneous Denoising and Hole Filling of Depth Image [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3473

ENHANCING PRODUCT IMAGES FOR CLICK-THROUGH RATE IMPROVEMENT


This paper proposes a statistical method to enhance image quality in order to increase the click-through rate (CTR) of product images. We build a joint probability model of global image features for photos of different product categories. The images are modified in terms of brightness, contrast, and sharpness in order to increase the expected CTR. The effectiveness of the method is evaluated using a perceptual user study, comparing it to histogram equalization methods, and by conducting an A/B test over one week on the e-commerce site Rakuten Ichiba.

Paper Details

Authors:
Yeongnam Chae, Mitsuru, Bjorn Stenger
Submitted On:
4 October 2018 - 9:03pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster_ICIP2018_fix.pdf

(32 downloads)

Subscribe

[1] Yeongnam Chae, Mitsuru, Bjorn Stenger, "ENHANCING PRODUCT IMAGES FOR CLICK-THROUGH RATE IMPROVEMENT", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3472. Accessed: Jan. 16, 2019.
@article{3472-18,
url = {http://sigport.org/3472},
author = {Yeongnam Chae; Mitsuru; Bjorn Stenger },
publisher = {IEEE SigPort},
title = {ENHANCING PRODUCT IMAGES FOR CLICK-THROUGH RATE IMPROVEMENT},
year = {2018} }
TY - EJOUR
T1 - ENHANCING PRODUCT IMAGES FOR CLICK-THROUGH RATE IMPROVEMENT
AU - Yeongnam Chae; Mitsuru; Bjorn Stenger
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3472
ER -
Yeongnam Chae, Mitsuru, Bjorn Stenger. (2018). ENHANCING PRODUCT IMAGES FOR CLICK-THROUGH RATE IMPROVEMENT. IEEE SigPort. http://sigport.org/3472
Yeongnam Chae, Mitsuru, Bjorn Stenger, 2018. ENHANCING PRODUCT IMAGES FOR CLICK-THROUGH RATE IMPROVEMENT. Available at: http://sigport.org/3472.
Yeongnam Chae, Mitsuru, Bjorn Stenger. (2018). "ENHANCING PRODUCT IMAGES FOR CLICK-THROUGH RATE IMPROVEMENT." Web.
1. Yeongnam Chae, Mitsuru, Bjorn Stenger. ENHANCING PRODUCT IMAGES FOR CLICK-THROUGH RATE IMPROVEMENT [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3472

Matching Pursuit Based on Kernel Non-second Order Minimization


The orthogonal matching pursuit (OMP) is an important sparse approximation algorithm to recover sparse signals from compressed measurements. However, most MP algorithms are based on the mean square error(MSE) to minimize the recovery error, which is suboptimal when there are outliers.

Paper Details

Authors:
Miaohua Zhang, Yongsheng Gao, Changming Sun, Michael Blumenstein
Submitted On:
4 October 2018 - 7:56pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Paper #3017.pdf

(36 downloads)

Keywords

Additional Categories

Subscribe

[1] Miaohua Zhang, Yongsheng Gao, Changming Sun, Michael Blumenstein, "Matching Pursuit Based on Kernel Non-second Order Minimization", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3470. Accessed: Jan. 16, 2019.
@article{3470-18,
url = {http://sigport.org/3470},
author = {Miaohua Zhang; Yongsheng Gao; Changming Sun; Michael Blumenstein },
publisher = {IEEE SigPort},
title = {Matching Pursuit Based on Kernel Non-second Order Minimization},
year = {2018} }
TY - EJOUR
T1 - Matching Pursuit Based on Kernel Non-second Order Minimization
AU - Miaohua Zhang; Yongsheng Gao; Changming Sun; Michael Blumenstein
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3470
ER -
Miaohua Zhang, Yongsheng Gao, Changming Sun, Michael Blumenstein. (2018). Matching Pursuit Based on Kernel Non-second Order Minimization. IEEE SigPort. http://sigport.org/3470
Miaohua Zhang, Yongsheng Gao, Changming Sun, Michael Blumenstein, 2018. Matching Pursuit Based on Kernel Non-second Order Minimization. Available at: http://sigport.org/3470.
Miaohua Zhang, Yongsheng Gao, Changming Sun, Michael Blumenstein. (2018). "Matching Pursuit Based on Kernel Non-second Order Minimization." Web.
1. Miaohua Zhang, Yongsheng Gao, Changming Sun, Michael Blumenstein. Matching Pursuit Based on Kernel Non-second Order Minimization [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3470

Pages