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

Deep Learning for Computer Vision

CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL


Occlusion removal is an interesting application of image enhancement, for which, existing work suggests manually-annotated or domain-specific occlusion removal. No work tries to address automatic occlusion detection and removal as a context-aware generic problem. In this paper, we present a novel methodology to identify objects that do not relate to the image context as occlusions and remove them, reconstructing the space occupied coherently.

Paper Details

Authors:
Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo
Submitted On:
21 September 2019 - 12:17am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL

(12)

Keywords

Additional Categories

Subscribe

[1] Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo, "CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4800. Accessed: Nov. 12, 2019.
@article{4800-19,
url = {http://sigport.org/4800},
author = {Kumara Kahatapitiya; Dumindu Tissera; Ranga Rodrigo },
publisher = {IEEE SigPort},
title = {CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL},
year = {2019} }
TY - EJOUR
T1 - CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL
AU - Kumara Kahatapitiya; Dumindu Tissera; Ranga Rodrigo
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4800
ER -
Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo. (2019). CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL. IEEE SigPort. http://sigport.org/4800
Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo, 2019. CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL. Available at: http://sigport.org/4800.
Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo. (2019). "CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL." Web.
1. Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo. CONTEXT-AWARE AUTOMATIC OCCLUSION REMOVAL [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4800

Deep Learning-based Obstacle Detection and Depth Estimation


This paper proposed a modified YOLOv3 which has an extra object depth prediction module for obstacle detection and avoidance. We use a pre-processed KITTI dataset to train the proposed, unified model for (i) object detection and (ii) depth prediction and use the AirSim flight simulator to generate synthetic aerial images to verify that our model can be applied in different data domains.

Paper Details

Authors:
Wei-Yu Lin, Dong-Lin Li, and Jen-Hui Chuang
Submitted On:
19 September 2019 - 9:32pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP_Deep Learning-based Obstacle Detection and Depth Estimation.pdf

(14)

Subscribe

[1] Wei-Yu Lin, Dong-Lin Li, and Jen-Hui Chuang, "Deep Learning-based Obstacle Detection and Depth Estimation", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4754. Accessed: Nov. 12, 2019.
@article{4754-19,
url = {http://sigport.org/4754},
author = {Wei-Yu Lin; Dong-Lin Li; and Jen-Hui Chuang },
publisher = {IEEE SigPort},
title = {Deep Learning-based Obstacle Detection and Depth Estimation},
year = {2019} }
TY - EJOUR
T1 - Deep Learning-based Obstacle Detection and Depth Estimation
AU - Wei-Yu Lin; Dong-Lin Li; and Jen-Hui Chuang
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4754
ER -
Wei-Yu Lin, Dong-Lin Li, and Jen-Hui Chuang. (2019). Deep Learning-based Obstacle Detection and Depth Estimation. IEEE SigPort. http://sigport.org/4754
Wei-Yu Lin, Dong-Lin Li, and Jen-Hui Chuang, 2019. Deep Learning-based Obstacle Detection and Depth Estimation. Available at: http://sigport.org/4754.
Wei-Yu Lin, Dong-Lin Li, and Jen-Hui Chuang. (2019). "Deep Learning-based Obstacle Detection and Depth Estimation." Web.
1. Wei-Yu Lin, Dong-Lin Li, and Jen-Hui Chuang. Deep Learning-based Obstacle Detection and Depth Estimation [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4754

ManGAN: Assisting Colorization of Manga Characters Concept Art using Conditional GAN


Colorization is a challenging task that has recently been tackled by deep learning. Line art colorization is particularly difficult because there is no grayscale value to indicate the color intensities as there is in black-and-white photograph images. When designing a character, concept artists often need to try different color schemes, however, colorization is a time-consuming task. In this article, we propose a semi-automatic framework for colorizing manga concept arts by letting concept artists try different color schemes and obtain colorized results in fashion time.

Paper Details

Authors:
Felipe Coelho Silva, Paulo André Lima de Castro, Hélio Ricardo Júnior, Ernesto Cordeiro Marujo
Submitted On:
18 September 2019 - 1:52pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ManGAN-good.pdf

(19)

Keywords

Additional Categories

Subscribe

[1] Felipe Coelho Silva, Paulo André Lima de Castro, Hélio Ricardo Júnior, Ernesto Cordeiro Marujo, "ManGAN: Assisting Colorization of Manga Characters Concept Art using Conditional GAN", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4692. Accessed: Nov. 12, 2019.
@article{4692-19,
url = {http://sigport.org/4692},
author = {Felipe Coelho Silva; Paulo André Lima de Castro; Hélio Ricardo Júnior; Ernesto Cordeiro Marujo },
publisher = {IEEE SigPort},
title = {ManGAN: Assisting Colorization of Manga Characters Concept Art using Conditional GAN},
year = {2019} }
TY - EJOUR
T1 - ManGAN: Assisting Colorization of Manga Characters Concept Art using Conditional GAN
AU - Felipe Coelho Silva; Paulo André Lima de Castro; Hélio Ricardo Júnior; Ernesto Cordeiro Marujo
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4692
ER -
Felipe Coelho Silva, Paulo André Lima de Castro, Hélio Ricardo Júnior, Ernesto Cordeiro Marujo. (2019). ManGAN: Assisting Colorization of Manga Characters Concept Art using Conditional GAN. IEEE SigPort. http://sigport.org/4692
Felipe Coelho Silva, Paulo André Lima de Castro, Hélio Ricardo Júnior, Ernesto Cordeiro Marujo, 2019. ManGAN: Assisting Colorization of Manga Characters Concept Art using Conditional GAN. Available at: http://sigport.org/4692.
Felipe Coelho Silva, Paulo André Lima de Castro, Hélio Ricardo Júnior, Ernesto Cordeiro Marujo. (2019). "ManGAN: Assisting Colorization of Manga Characters Concept Art using Conditional GAN." Web.
1. Felipe Coelho Silva, Paulo André Lima de Castro, Hélio Ricardo Júnior, Ernesto Cordeiro Marujo. ManGAN: Assisting Colorization of Manga Characters Concept Art using Conditional GAN [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4692

Context-Aware Automatic Occlusion Removal


Occlusion removal is an interesting application of image enhancement, for which, existing work suggests manually-

Paper Details

Authors:
Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo
Submitted On:
20 September 2019 - 9:19am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

context aware automatic occlusion removal

(14)

ICIP_poster.pdf

(9)

Keywords

Additional Categories

Subscribe

[1] Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo, "Context-Aware Automatic Occlusion Removal", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4667. Accessed: Nov. 12, 2019.
@article{4667-19,
url = {http://sigport.org/4667},
author = {Kumara Kahatapitiya; Dumindu Tissera; Ranga Rodrigo },
publisher = {IEEE SigPort},
title = {Context-Aware Automatic Occlusion Removal},
year = {2019} }
TY - EJOUR
T1 - Context-Aware Automatic Occlusion Removal
AU - Kumara Kahatapitiya; Dumindu Tissera; Ranga Rodrigo
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4667
ER -
Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo. (2019). Context-Aware Automatic Occlusion Removal. IEEE SigPort. http://sigport.org/4667
Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo, 2019. Context-Aware Automatic Occlusion Removal. Available at: http://sigport.org/4667.
Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo. (2019). "Context-Aware Automatic Occlusion Removal." Web.
1. Kumara Kahatapitiya, Dumindu Tissera, Ranga Rodrigo. Context-Aware Automatic Occlusion Removal [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4667

TRAINING ACCURATE BINARY NEURAL NETWORKS FROM SCRATCH


Binary neural networks are a promising approach to execute convolutional neural networks on devices with low computational power. Previous work on this subject often quantizes pretrained full-precision models and uses complex training strategies. In our work, we focus on increasing the performance of binary neural networks by training from scratch with a simple training strategy. In our experiments we show that we are able to achieve state-of-the-art results on standard benchmark datasets.

Paper Details

Authors:
Joseph Bethge, Haojin Yang, Christoph Meinel
Submitted On:
11 September 2019 - 8:14am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

poster.pdf

(18)

Keywords

Additional Categories

Subscribe

[1] Joseph Bethge, Haojin Yang, Christoph Meinel, "TRAINING ACCURATE BINARY NEURAL NETWORKS FROM SCRATCH", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4590. Accessed: Nov. 12, 2019.
@article{4590-19,
url = {http://sigport.org/4590},
author = {Joseph Bethge; Haojin Yang; Christoph Meinel },
publisher = {IEEE SigPort},
title = {TRAINING ACCURATE BINARY NEURAL NETWORKS FROM SCRATCH},
year = {2019} }
TY - EJOUR
T1 - TRAINING ACCURATE BINARY NEURAL NETWORKS FROM SCRATCH
AU - Joseph Bethge; Haojin Yang; Christoph Meinel
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4590
ER -
Joseph Bethge, Haojin Yang, Christoph Meinel. (2019). TRAINING ACCURATE BINARY NEURAL NETWORKS FROM SCRATCH. IEEE SigPort. http://sigport.org/4590
Joseph Bethge, Haojin Yang, Christoph Meinel, 2019. TRAINING ACCURATE BINARY NEURAL NETWORKS FROM SCRATCH. Available at: http://sigport.org/4590.
Joseph Bethge, Haojin Yang, Christoph Meinel. (2019). "TRAINING ACCURATE BINARY NEURAL NETWORKS FROM SCRATCH." Web.
1. Joseph Bethge, Haojin Yang, Christoph Meinel. TRAINING ACCURATE BINARY NEURAL NETWORKS FROM SCRATCH [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4590

LEARNING TEMPORAL INFORMATION FROM SPATIAL INFORMATION USING CAPSNETS FOR HUMAN ACTION RECOGNITION


Capsule Networks (CapsNets) are recently introduced to overcome some of the shortcomings of traditional Convolutional Neural Networks (CNNs). CapsNets replace neurons in CNNs with vectors to retain spatial relationships among the features. In this paper, we propose a CapsNet architecture that employs individual video frames for human action recognition without explicitly extracting motion information. We also propose weight pooling to reduce the computational complexity and improve the classification accuracy by appropriately removing some of the extracted features.

Paper Details

Authors:
Abdullah M. Algamdi , Victor Sanchez , Chang-Tsun Li
Submitted On:
8 May 2019 - 8:03am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP_poster_2019__1_ (2).pdf

(33)

Keywords

Additional Categories

Subscribe

[1] Abdullah M. Algamdi , Victor Sanchez , Chang-Tsun Li, "LEARNING TEMPORAL INFORMATION FROM SPATIAL INFORMATION USING CAPSNETS FOR HUMAN ACTION RECOGNITION", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4079. Accessed: Nov. 12, 2019.
@article{4079-19,
url = {http://sigport.org/4079},
author = {Abdullah M. Algamdi ; Victor Sanchez ; Chang-Tsun Li },
publisher = {IEEE SigPort},
title = {LEARNING TEMPORAL INFORMATION FROM SPATIAL INFORMATION USING CAPSNETS FOR HUMAN ACTION RECOGNITION},
year = {2019} }
TY - EJOUR
T1 - LEARNING TEMPORAL INFORMATION FROM SPATIAL INFORMATION USING CAPSNETS FOR HUMAN ACTION RECOGNITION
AU - Abdullah M. Algamdi ; Victor Sanchez ; Chang-Tsun Li
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4079
ER -
Abdullah M. Algamdi , Victor Sanchez , Chang-Tsun Li. (2019). LEARNING TEMPORAL INFORMATION FROM SPATIAL INFORMATION USING CAPSNETS FOR HUMAN ACTION RECOGNITION. IEEE SigPort. http://sigport.org/4079
Abdullah M. Algamdi , Victor Sanchez , Chang-Tsun Li, 2019. LEARNING TEMPORAL INFORMATION FROM SPATIAL INFORMATION USING CAPSNETS FOR HUMAN ACTION RECOGNITION. Available at: http://sigport.org/4079.
Abdullah M. Algamdi , Victor Sanchez , Chang-Tsun Li. (2019). "LEARNING TEMPORAL INFORMATION FROM SPATIAL INFORMATION USING CAPSNETS FOR HUMAN ACTION RECOGNITION." Web.
1. Abdullah M. Algamdi , Victor Sanchez , Chang-Tsun Li. LEARNING TEMPORAL INFORMATION FROM SPATIAL INFORMATION USING CAPSNETS FOR HUMAN ACTION RECOGNITION [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4079

RTSeg: Real-time Semantic Segmentation Comparative Study

Paper Details

Authors:
Submitted On:
7 October 2018 - 3:57pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP_Poster.pdf

(359)

Keywords

Additional Categories

Subscribe

[1] , "RTSeg: Real-time Semantic Segmentation Comparative Study", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3600. Accessed: Nov. 12, 2019.
@article{3600-18,
url = {http://sigport.org/3600},
author = { },
publisher = {IEEE SigPort},
title = {RTSeg: Real-time Semantic Segmentation Comparative Study},
year = {2018} }
TY - EJOUR
T1 - RTSeg: Real-time Semantic Segmentation Comparative Study
AU -
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3600
ER -
. (2018). RTSeg: Real-time Semantic Segmentation Comparative Study. IEEE SigPort. http://sigport.org/3600
, 2018. RTSeg: Real-time Semantic Segmentation Comparative Study. Available at: http://sigport.org/3600.
. (2018). "RTSeg: Real-time Semantic Segmentation Comparative Study." Web.
1. . RTSeg: Real-time Semantic Segmentation Comparative Study [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3600

A-CCNN: Adaptive CCNN for Density Estimation and Crowd Counting


Crowd counting, for estimating the number of people in a crowd using vision-based computer techniques, has attracted much interest in the research community. Although many attempts have been reported, real-world problems, such as huge variation in subjects’ sizes in images and serious occlusion among people, make it still a challenging problem. In this paper, we propose an Adaptive Counting Convolutional Neural Network (A-CCNN) and consider the scale variation of objects in a frame adaptively so as to improve the accuracy of counting.

Paper Details

Authors:
Saeed Amirgholipour, Xiangjian He, Wenjing Jia, Dadong Wang, Michelle Zeibots
Submitted On:
10 October 2018 - 7:26am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2018.pptx

(68)

Keywords

Additional Categories

Subscribe

[1] Saeed Amirgholipour, Xiangjian He, Wenjing Jia, Dadong Wang, Michelle Zeibots, "A-CCNN: Adaptive CCNN for Density Estimation and Crowd Counting", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3599. Accessed: Nov. 12, 2019.
@article{3599-18,
url = {http://sigport.org/3599},
author = {Saeed Amirgholipour; Xiangjian He; Wenjing Jia; Dadong Wang; Michelle Zeibots },
publisher = {IEEE SigPort},
title = {A-CCNN: Adaptive CCNN for Density Estimation and Crowd Counting},
year = {2018} }
TY - EJOUR
T1 - A-CCNN: Adaptive CCNN for Density Estimation and Crowd Counting
AU - Saeed Amirgholipour; Xiangjian He; Wenjing Jia; Dadong Wang; Michelle Zeibots
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3599
ER -
Saeed Amirgholipour, Xiangjian He, Wenjing Jia, Dadong Wang, Michelle Zeibots. (2018). A-CCNN: Adaptive CCNN for Density Estimation and Crowd Counting. IEEE SigPort. http://sigport.org/3599
Saeed Amirgholipour, Xiangjian He, Wenjing Jia, Dadong Wang, Michelle Zeibots, 2018. A-CCNN: Adaptive CCNN for Density Estimation and Crowd Counting. Available at: http://sigport.org/3599.
Saeed Amirgholipour, Xiangjian He, Wenjing Jia, Dadong Wang, Michelle Zeibots. (2018). "A-CCNN: Adaptive CCNN for Density Estimation and Crowd Counting." Web.
1. Saeed Amirgholipour, Xiangjian He, Wenjing Jia, Dadong Wang, Michelle Zeibots. A-CCNN: Adaptive CCNN for Density Estimation and Crowd Counting [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3599

Deep Feature Embedding Learning for Person Re-Identification Using Lifted Structured Loss


In this paper, we propose deep feature embedding learning for person re-identification (re-id) using lifted structured loss. Although triplet loss has been commonly used in deep neural networks for person re-id, the triplet loss-based framework is not effective in fully using the batch information. Thus, it needs to choose hard negative samples manually that is very time-consuming. To address this problem, we adopt lifted structured loss for deep neural networks that makes the network learn better feature embedding by minimizing intra-class variation and maximizing inter-class variation.

Paper Details

Authors:
Zhangping He, Zhendong Zhang, Cheolkon Jung
Submitted On:
20 April 2018 - 5:23am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

ICASSP2018_PersonReID_final.pdf

(193)

Keywords

Additional Categories

Subscribe

[1] Zhangping He, Zhendong Zhang, Cheolkon Jung, "Deep Feature Embedding Learning for Person Re-Identification Using Lifted Structured Loss", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3098. Accessed: Nov. 12, 2019.
@article{3098-18,
url = {http://sigport.org/3098},
author = {Zhangping He; Zhendong Zhang; Cheolkon Jung },
publisher = {IEEE SigPort},
title = {Deep Feature Embedding Learning for Person Re-Identification Using Lifted Structured Loss},
year = {2018} }
TY - EJOUR
T1 - Deep Feature Embedding Learning for Person Re-Identification Using Lifted Structured Loss
AU - Zhangping He; Zhendong Zhang; Cheolkon Jung
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3098
ER -
Zhangping He, Zhendong Zhang, Cheolkon Jung. (2018). Deep Feature Embedding Learning for Person Re-Identification Using Lifted Structured Loss. IEEE SigPort. http://sigport.org/3098
Zhangping He, Zhendong Zhang, Cheolkon Jung, 2018. Deep Feature Embedding Learning for Person Re-Identification Using Lifted Structured Loss. Available at: http://sigport.org/3098.
Zhangping He, Zhendong Zhang, Cheolkon Jung. (2018). "Deep Feature Embedding Learning for Person Re-Identification Using Lifted Structured Loss." Web.
1. Zhangping He, Zhendong Zhang, Cheolkon Jung. Deep Feature Embedding Learning for Person Re-Identification Using Lifted Structured Loss [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3098