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

Image/Video Storage, Retrieval

A FEATURE EMBEDDING STRATEGY FOR HIGH-LEVEL CNN REPRESENTATIONS FROM MULTIPLE CONVNETS

Paper Details

Authors:
Thangarajah Akilan, Jonathan Wu, Wei Jiang
Submitted On:
13 November 2017 - 9:13pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Transfer learning and feature embedding

(23 downloads)

Keywords

Additional Categories

Subscribe

[1] Thangarajah Akilan, Jonathan Wu, Wei Jiang, " A FEATURE EMBEDDING STRATEGY FOR HIGH-LEVEL CNN REPRESENTATIONS FROM MULTIPLE CONVNETS", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2345. Accessed: Dec. 17, 2017.
@article{2345-17,
url = {http://sigport.org/2345},
author = {Thangarajah Akilan; Jonathan Wu; Wei Jiang },
publisher = {IEEE SigPort},
title = { A FEATURE EMBEDDING STRATEGY FOR HIGH-LEVEL CNN REPRESENTATIONS FROM MULTIPLE CONVNETS},
year = {2017} }
TY - EJOUR
T1 - A FEATURE EMBEDDING STRATEGY FOR HIGH-LEVEL CNN REPRESENTATIONS FROM MULTIPLE CONVNETS
AU - Thangarajah Akilan; Jonathan Wu; Wei Jiang
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2345
ER -
Thangarajah Akilan, Jonathan Wu, Wei Jiang. (2017). A FEATURE EMBEDDING STRATEGY FOR HIGH-LEVEL CNN REPRESENTATIONS FROM MULTIPLE CONVNETS. IEEE SigPort. http://sigport.org/2345
Thangarajah Akilan, Jonathan Wu, Wei Jiang, 2017. A FEATURE EMBEDDING STRATEGY FOR HIGH-LEVEL CNN REPRESENTATIONS FROM MULTIPLE CONVNETS. Available at: http://sigport.org/2345.
Thangarajah Akilan, Jonathan Wu, Wei Jiang. (2017). " A FEATURE EMBEDDING STRATEGY FOR HIGH-LEVEL CNN REPRESENTATIONS FROM MULTIPLE CONVNETS." Web.
1. Thangarajah Akilan, Jonathan Wu, Wei Jiang. A FEATURE EMBEDDING STRATEGY FOR HIGH-LEVEL CNN REPRESENTATIONS FROM MULTIPLE CONVNETS [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2345

End-to-end Learning Binary Representation via Direct Binary Embedding


Learning binary representation is essential to large-scale computer vision tasks. Most existing algorithms require a separate quantization constraint to learn effective hashing functions. In this work, we present Direct Binary Embedding (DBE), a simple yet very effective algorithm to learn binary representation in an end-to-end fashion. By appending an ingeniously designed DBE layer to the deep convolutional neural network (DCNN), DBE learns binary code directly from the continuous DBE layer activation without quantization error.

Paper Details

Authors:
Hairong Qi
Submitted On:
16 September 2017 - 12:01pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Liu_icip_17.pdf

(36 downloads)

Keywords

Subscribe

[1] Hairong Qi, "End-to-end Learning Binary Representation via Direct Binary Embedding", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2199. Accessed: Dec. 17, 2017.
@article{2199-17,
url = {http://sigport.org/2199},
author = {Hairong Qi },
publisher = {IEEE SigPort},
title = {End-to-end Learning Binary Representation via Direct Binary Embedding},
year = {2017} }
TY - EJOUR
T1 - End-to-end Learning Binary Representation via Direct Binary Embedding
AU - Hairong Qi
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2199
ER -
Hairong Qi. (2017). End-to-end Learning Binary Representation via Direct Binary Embedding. IEEE SigPort. http://sigport.org/2199
Hairong Qi, 2017. End-to-end Learning Binary Representation via Direct Binary Embedding. Available at: http://sigport.org/2199.
Hairong Qi. (2017). "End-to-end Learning Binary Representation via Direct Binary Embedding." Web.
1. Hairong Qi. End-to-end Learning Binary Representation via Direct Binary Embedding [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2199

Slides for ID 2913 at ICIP 2017

Paper Details

Authors:
Junfu Pu, Yusuke Matsui, Fan Yang, Shin'ichi Satoh
Submitted On:
15 September 2017 - 8:28pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2017_ID2913.pdf

(60 downloads)

Keywords

Subscribe

[1] Junfu Pu, Yusuke Matsui, Fan Yang, Shin'ichi Satoh, "Slides for ID 2913 at ICIP 2017", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2163. Accessed: Dec. 17, 2017.
@article{2163-17,
url = {http://sigport.org/2163},
author = {Junfu Pu; Yusuke Matsui; Fan Yang; Shin'ichi Satoh },
publisher = {IEEE SigPort},
title = {Slides for ID 2913 at ICIP 2017},
year = {2017} }
TY - EJOUR
T1 - Slides for ID 2913 at ICIP 2017
AU - Junfu Pu; Yusuke Matsui; Fan Yang; Shin'ichi Satoh
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2163
ER -
Junfu Pu, Yusuke Matsui, Fan Yang, Shin'ichi Satoh. (2017). Slides for ID 2913 at ICIP 2017. IEEE SigPort. http://sigport.org/2163
Junfu Pu, Yusuke Matsui, Fan Yang, Shin'ichi Satoh, 2017. Slides for ID 2913 at ICIP 2017. Available at: http://sigport.org/2163.
Junfu Pu, Yusuke Matsui, Fan Yang, Shin'ichi Satoh. (2017). "Slides for ID 2913 at ICIP 2017." Web.
1. Junfu Pu, Yusuke Matsui, Fan Yang, Shin'ichi Satoh. Slides for ID 2913 at ICIP 2017 [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2163

LEARNING CIRCULANT SUPPORT VECTOR MACHINES FOR FAST IMAGE SEARCH


Binary hashing is an established approach for fast, approximate image search. It maps a query image to a binary vector so that Hamming distances approximate image similarities. Applying the hash function can be made fast by using a circulant matrix and the fast Fourier transform, but this circulant hash function must be learned optimally from training data. We show that a previously proposed learning algorithm based on optimization in the frequency domain is suboptimal.

Paper Details

Authors:
Ramin Raziperchikolaei, Miguel Carreira-Perpinan
Submitted On:
15 September 2017 - 7:36pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

icip17b-slides.pdf

(34 downloads)

Keywords

Subscribe

[1] Ramin Raziperchikolaei, Miguel Carreira-Perpinan, "LEARNING CIRCULANT SUPPORT VECTOR MACHINES FOR FAST IMAGE SEARCH", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2162. Accessed: Dec. 17, 2017.
@article{2162-17,
url = {http://sigport.org/2162},
author = {Ramin Raziperchikolaei; Miguel Carreira-Perpinan },
publisher = {IEEE SigPort},
title = {LEARNING CIRCULANT SUPPORT VECTOR MACHINES FOR FAST IMAGE SEARCH},
year = {2017} }
TY - EJOUR
T1 - LEARNING CIRCULANT SUPPORT VECTOR MACHINES FOR FAST IMAGE SEARCH
AU - Ramin Raziperchikolaei; Miguel Carreira-Perpinan
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2162
ER -
Ramin Raziperchikolaei, Miguel Carreira-Perpinan. (2017). LEARNING CIRCULANT SUPPORT VECTOR MACHINES FOR FAST IMAGE SEARCH. IEEE SigPort. http://sigport.org/2162
Ramin Raziperchikolaei, Miguel Carreira-Perpinan, 2017. LEARNING CIRCULANT SUPPORT VECTOR MACHINES FOR FAST IMAGE SEARCH. Available at: http://sigport.org/2162.
Ramin Raziperchikolaei, Miguel Carreira-Perpinan. (2017). "LEARNING CIRCULANT SUPPORT VECTOR MACHINES FOR FAST IMAGE SEARCH." Web.
1. Ramin Raziperchikolaei, Miguel Carreira-Perpinan. LEARNING CIRCULANT SUPPORT VECTOR MACHINES FOR FAST IMAGE SEARCH [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2162

ICIP2017_quasi rate distortion optimization for binary hashing

Paper Details

Authors:
Yiding Liu, Wengang Zhou, and Houqiang Li
Submitted On:
15 September 2017 - 4:54pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP17_Quasi Rate Distortion Optimization for Binary Hashing.pdf

(29 downloads)

Keywords

Subscribe

[1] Yiding Liu, Wengang Zhou, and Houqiang Li, "ICIP2017_quasi rate distortion optimization for binary hashing", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2160. Accessed: Dec. 17, 2017.
@article{2160-17,
url = {http://sigport.org/2160},
author = {Yiding Liu; Wengang Zhou; and Houqiang Li },
publisher = {IEEE SigPort},
title = {ICIP2017_quasi rate distortion optimization for binary hashing},
year = {2017} }
TY - EJOUR
T1 - ICIP2017_quasi rate distortion optimization for binary hashing
AU - Yiding Liu; Wengang Zhou; and Houqiang Li
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2160
ER -
Yiding Liu, Wengang Zhou, and Houqiang Li. (2017). ICIP2017_quasi rate distortion optimization for binary hashing. IEEE SigPort. http://sigport.org/2160
Yiding Liu, Wengang Zhou, and Houqiang Li, 2017. ICIP2017_quasi rate distortion optimization for binary hashing. Available at: http://sigport.org/2160.
Yiding Liu, Wengang Zhou, and Houqiang Li. (2017). "ICIP2017_quasi rate distortion optimization for binary hashing." Web.
1. Yiding Liu, Wengang Zhou, and Houqiang Li. ICIP2017_quasi rate distortion optimization for binary hashing [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2160

SUPERVISED HASHING WITH JOINTLY LEARNING EMBEDDING AND QUANTIZATION


Compared with unsupervised hashing, supervised hashing commonly illustrates better accuracy in many real applica- tions by leveraging semantic (label) information. However, it is tough to solve the supervised hashing problem directly because it is essentially a discrete optimization problem. Some other works try to solve the discrete optimization problem directly using binary quadratic programming, but they are typically too complicated and time-consuming while some supervised hashing methods have to solve a relaxed continuous optimization problem by dropping the discrete con- straints.

ICIP3325.pdf

PDF icon ICIP3325.pdf (33 downloads)

Paper Details

Authors:
Xiang Xiang;Feng Wang;Trac D. Tran
Submitted On:
15 September 2017 - 11:51am
Short Link:
Type:
Event:
Paper Code:
Document Year:
Cite

Document Files

ICIP3325.pdf

(33 downloads)

Keywords

Subscribe

[1] Xiang Xiang;Feng Wang;Trac D. Tran, "SUPERVISED HASHING WITH JOINTLY LEARNING EMBEDDING AND QUANTIZATION", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2147. Accessed: Dec. 17, 2017.
@article{2147-17,
url = {http://sigport.org/2147},
author = {Xiang Xiang;Feng Wang;Trac D. Tran },
publisher = {IEEE SigPort},
title = {SUPERVISED HASHING WITH JOINTLY LEARNING EMBEDDING AND QUANTIZATION},
year = {2017} }
TY - EJOUR
T1 - SUPERVISED HASHING WITH JOINTLY LEARNING EMBEDDING AND QUANTIZATION
AU - Xiang Xiang;Feng Wang;Trac D. Tran
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2147
ER -
Xiang Xiang;Feng Wang;Trac D. Tran. (2017). SUPERVISED HASHING WITH JOINTLY LEARNING EMBEDDING AND QUANTIZATION. IEEE SigPort. http://sigport.org/2147
Xiang Xiang;Feng Wang;Trac D. Tran, 2017. SUPERVISED HASHING WITH JOINTLY LEARNING EMBEDDING AND QUANTIZATION. Available at: http://sigport.org/2147.
Xiang Xiang;Feng Wang;Trac D. Tran. (2017). "SUPERVISED HASHING WITH JOINTLY LEARNING EMBEDDING AND QUANTIZATION." Web.
1. Xiang Xiang;Feng Wang;Trac D. Tran. SUPERVISED HASHING WITH JOINTLY LEARNING EMBEDDING AND QUANTIZATION [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2147

An Object Based Graph Representation for Video Comparison


This paper develops a novel object based graph model for semantic video comparison. The model describes a video with detected objects as nodes, and elationship between the objects as edges in a graph. We investigated several spatial and temporal features as the graph node attributes, and dierent ways to describe the spatial-temporal relationship between objects as the edge attributes. To tackle the problem of erratic camera motion on the detected object, a global motion estimation and correction approach is proposed to reveal the true object trajectory.

Paper Details

Authors:
Xin Feng, Yuanyi Xue, Yao Wang
Submitted On:
15 September 2017 - 11:50am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

VOG_poster_print_revised.pdf

(29 downloads)

Keywords

Subscribe

[1] Xin Feng, Yuanyi Xue, Yao Wang, "An Object Based Graph Representation for Video Comparison", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2145. Accessed: Dec. 17, 2017.
@article{2145-17,
url = {http://sigport.org/2145},
author = {Xin Feng; Yuanyi Xue; Yao Wang },
publisher = {IEEE SigPort},
title = {An Object Based Graph Representation for Video Comparison},
year = {2017} }
TY - EJOUR
T1 - An Object Based Graph Representation for Video Comparison
AU - Xin Feng; Yuanyi Xue; Yao Wang
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2145
ER -
Xin Feng, Yuanyi Xue, Yao Wang. (2017). An Object Based Graph Representation for Video Comparison. IEEE SigPort. http://sigport.org/2145
Xin Feng, Yuanyi Xue, Yao Wang, 2017. An Object Based Graph Representation for Video Comparison. Available at: http://sigport.org/2145.
Xin Feng, Yuanyi Xue, Yao Wang. (2017). "An Object Based Graph Representation for Video Comparison." Web.
1. Xin Feng, Yuanyi Xue, Yao Wang. An Object Based Graph Representation for Video Comparison [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2145

ACTION RECOGNITION USING SPATIO-TEMPORAL DIFFERENTIAL MOTION

Paper Details

Authors:
Gaurav Kumar Yadav, Amit Sethi
Submitted On:
15 September 2017 - 6:09am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Gaurav_poster.pdf

(26 downloads)

Keywords

Subscribe

[1] Gaurav Kumar Yadav, Amit Sethi, "ACTION RECOGNITION USING SPATIO-TEMPORAL DIFFERENTIAL MOTION", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2116. Accessed: Dec. 17, 2017.
@article{2116-17,
url = {http://sigport.org/2116},
author = {Gaurav Kumar Yadav; Amit Sethi },
publisher = {IEEE SigPort},
title = {ACTION RECOGNITION USING SPATIO-TEMPORAL DIFFERENTIAL MOTION},
year = {2017} }
TY - EJOUR
T1 - ACTION RECOGNITION USING SPATIO-TEMPORAL DIFFERENTIAL MOTION
AU - Gaurav Kumar Yadav; Amit Sethi
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2116
ER -
Gaurav Kumar Yadav, Amit Sethi. (2017). ACTION RECOGNITION USING SPATIO-TEMPORAL DIFFERENTIAL MOTION. IEEE SigPort. http://sigport.org/2116
Gaurav Kumar Yadav, Amit Sethi, 2017. ACTION RECOGNITION USING SPATIO-TEMPORAL DIFFERENTIAL MOTION. Available at: http://sigport.org/2116.
Gaurav Kumar Yadav, Amit Sethi. (2017). "ACTION RECOGNITION USING SPATIO-TEMPORAL DIFFERENTIAL MOTION." Web.
1. Gaurav Kumar Yadav, Amit Sethi. ACTION RECOGNITION USING SPATIO-TEMPORAL DIFFERENTIAL MOTION [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2116

Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network


In this paper, we aim to find exactly the same shoes given a daily shoe photo (street scenario) that matches the online shop shoe photo (shop scenario). There are large visual differences between the street and shop scenario shoe images. To handle the discrepancy of different scenarios, we learn a feature embedding for shoes via a viewpoint-invariant triplet network, the feature activations of which reflect the inherent similarity between any two shoe images.

Paper Details

Authors:
Huijing Zhan, Boxin Shi, Alex C. Kot
Submitted On:
15 September 2017 - 1:04am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2017_V6.pptx

(28 downloads)

Keywords

Subscribe

[1] Huijing Zhan, Boxin Shi, Alex C. Kot, "Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2082. Accessed: Dec. 17, 2017.
@article{2082-17,
url = {http://sigport.org/2082},
author = {Huijing Zhan; Boxin Shi; Alex C. Kot },
publisher = {IEEE SigPort},
title = {Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network},
year = {2017} }
TY - EJOUR
T1 - Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network
AU - Huijing Zhan; Boxin Shi; Alex C. Kot
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2082
ER -
Huijing Zhan, Boxin Shi, Alex C. Kot. (2017). Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network. IEEE SigPort. http://sigport.org/2082
Huijing Zhan, Boxin Shi, Alex C. Kot, 2017. Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network. Available at: http://sigport.org/2082.
Huijing Zhan, Boxin Shi, Alex C. Kot. (2017). "Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network." Web.
1. Huijing Zhan, Boxin Shi, Alex C. Kot. Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2082

Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network


In this paper we aim to find exactly the same shoes given a daily shoe photo (street scenario) that matches the online shop shoe photo (shop scenario). There are large visual differences between the street and shop scenario shoe images. To handle the discrepancy of different scenarios, we learn a feature embedding for shoes via a viewpoint-invariant triplet network, the feature activations of which reflect the inherent similarity between any two shoe images.

Paper Details

Authors:
Huijing Zhan, Boxin Shi, Alex C. Kot
Submitted On:
15 September 2017 - 12:50am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2017_V6.pptx

(38 downloads)

Keywords

Subscribe

[1] Huijing Zhan, Boxin Shi, Alex C. Kot, "Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2081. Accessed: Dec. 17, 2017.
@article{2081-17,
url = {http://sigport.org/2081},
author = {Huijing Zhan; Boxin Shi; Alex C. Kot },
publisher = {IEEE SigPort},
title = {Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network},
year = {2017} }
TY - EJOUR
T1 - Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network
AU - Huijing Zhan; Boxin Shi; Alex C. Kot
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2081
ER -
Huijing Zhan, Boxin Shi, Alex C. Kot. (2017). Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network. IEEE SigPort. http://sigport.org/2081
Huijing Zhan, Boxin Shi, Alex C. Kot, 2017. Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network. Available at: http://sigport.org/2081.
Huijing Zhan, Boxin Shi, Alex C. Kot. (2017). "Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network." Web.
1. Huijing Zhan, Boxin Shi, Alex C. Kot. Street-to-Shop Shoe Retrieval with Multi-Scale Viewpoint Invariant Triplet Network [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2081

Pages