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

Machine Learning for Signal Processing

TOWARDS 3D CONVOLUTIONAL NEURAL NETWORKS WITH MESHES


Voxels are an effective approach to 3D mesh and point cloud classification because they build upon mature Convolutional Neural Network concepts. We show however that their cubic increase in dimensionality is unsuitable for more challenging problems such as object detection in a complex point cloud scene. We observe that 3D meshes are analogous to graph data and can thus be treated with graph signal processing techniques.

Paper Details

Authors:
Felipe Petroski Such, Shagan Sah, Raymond Ptucha
Submitted On:
19 September 2017 - 11:34am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIPPoster2017MiguelDominguez.pdf

(2 downloads)

Keywords

Subscribe

[1] Felipe Petroski Such, Shagan Sah, Raymond Ptucha, "TOWARDS 3D CONVOLUTIONAL NEURAL NETWORKS WITH MESHES", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2234. Accessed: Sep. 20, 2017.
@article{2234-17,
url = {http://sigport.org/2234},
author = {Felipe Petroski Such; Shagan Sah; Raymond Ptucha },
publisher = {IEEE SigPort},
title = {TOWARDS 3D CONVOLUTIONAL NEURAL NETWORKS WITH MESHES},
year = {2017} }
TY - EJOUR
T1 - TOWARDS 3D CONVOLUTIONAL NEURAL NETWORKS WITH MESHES
AU - Felipe Petroski Such; Shagan Sah; Raymond Ptucha
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2234
ER -
Felipe Petroski Such, Shagan Sah, Raymond Ptucha. (2017). TOWARDS 3D CONVOLUTIONAL NEURAL NETWORKS WITH MESHES. IEEE SigPort. http://sigport.org/2234
Felipe Petroski Such, Shagan Sah, Raymond Ptucha, 2017. TOWARDS 3D CONVOLUTIONAL NEURAL NETWORKS WITH MESHES. Available at: http://sigport.org/2234.
Felipe Petroski Such, Shagan Sah, Raymond Ptucha. (2017). "TOWARDS 3D CONVOLUTIONAL NEURAL NETWORKS WITH MESHES." Web.
1. Felipe Petroski Such, Shagan Sah, Raymond Ptucha. TOWARDS 3D CONVOLUTIONAL NEURAL NETWORKS WITH MESHES [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2234

Greedy Deep Transform Learning


We introduce deep transform learning – a new
tool for deep learning. Deeper representation is learnt by
stacking one transform after another. The learning proceeds in
a greedy way. The first layer learns the transform and features
from the input training samples. Subsequent layers use the
features (after activation) from the previous layers as training
input. Experiments have been carried out with other deep
representation learning tools – deep dictionary learning,
stacked denoising autoencoder, deep belief network and PCANet

Paper Details

Authors:
Jyoti Maggu, Angshul Majumdar
Submitted On:
18 September 2017 - 1:57pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP_greedyDTL.pdf

(8 downloads)

Keywords

Subscribe

[1] Jyoti Maggu, Angshul Majumdar, "Greedy Deep Transform Learning", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2180. Accessed: Sep. 20, 2017.
@article{2180-17,
url = {http://sigport.org/2180},
author = {Jyoti Maggu; Angshul Majumdar },
publisher = {IEEE SigPort},
title = {Greedy Deep Transform Learning},
year = {2017} }
TY - EJOUR
T1 - Greedy Deep Transform Learning
AU - Jyoti Maggu; Angshul Majumdar
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2180
ER -
Jyoti Maggu, Angshul Majumdar. (2017). Greedy Deep Transform Learning. IEEE SigPort. http://sigport.org/2180
Jyoti Maggu, Angshul Majumdar, 2017. Greedy Deep Transform Learning. Available at: http://sigport.org/2180.
Jyoti Maggu, Angshul Majumdar. (2017). "Greedy Deep Transform Learning." Web.
1. Jyoti Maggu, Angshul Majumdar. Greedy Deep Transform Learning [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2180

AN IMAGE RECONSTRUCTION FRAMEWORK BASED ON DEEP NEURAL NETWORK FOR ELECTRICAL IMPEDANCE TOMOGRAPHY

Paper Details

Authors:
Xiuyan Li, Yang Lu,Jianming Wang,Xin Dang,Qi Wang,Xiaojie Duan,Yukuan Sun
Submitted On:
15 September 2017 - 12:21am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster for paper 3030-ICIP2017.pdf

(6 downloads)

Keywords

Subscribe

[1] Xiuyan Li, Yang Lu,Jianming Wang,Xin Dang,Qi Wang,Xiaojie Duan,Yukuan Sun, "AN IMAGE RECONSTRUCTION FRAMEWORK BASED ON DEEP NEURAL NETWORK FOR ELECTRICAL IMPEDANCE TOMOGRAPHY", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2078. Accessed: Sep. 20, 2017.
@article{2078-17,
url = {http://sigport.org/2078},
author = {Xiuyan Li; Yang Lu,Jianming Wang,Xin Dang,Qi Wang,Xiaojie Duan,Yukuan Sun },
publisher = {IEEE SigPort},
title = {AN IMAGE RECONSTRUCTION FRAMEWORK BASED ON DEEP NEURAL NETWORK FOR ELECTRICAL IMPEDANCE TOMOGRAPHY},
year = {2017} }
TY - EJOUR
T1 - AN IMAGE RECONSTRUCTION FRAMEWORK BASED ON DEEP NEURAL NETWORK FOR ELECTRICAL IMPEDANCE TOMOGRAPHY
AU - Xiuyan Li; Yang Lu,Jianming Wang,Xin Dang,Qi Wang,Xiaojie Duan,Yukuan Sun
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2078
ER -
Xiuyan Li, Yang Lu,Jianming Wang,Xin Dang,Qi Wang,Xiaojie Duan,Yukuan Sun. (2017). AN IMAGE RECONSTRUCTION FRAMEWORK BASED ON DEEP NEURAL NETWORK FOR ELECTRICAL IMPEDANCE TOMOGRAPHY. IEEE SigPort. http://sigport.org/2078
Xiuyan Li, Yang Lu,Jianming Wang,Xin Dang,Qi Wang,Xiaojie Duan,Yukuan Sun, 2017. AN IMAGE RECONSTRUCTION FRAMEWORK BASED ON DEEP NEURAL NETWORK FOR ELECTRICAL IMPEDANCE TOMOGRAPHY. Available at: http://sigport.org/2078.
Xiuyan Li, Yang Lu,Jianming Wang,Xin Dang,Qi Wang,Xiaojie Duan,Yukuan Sun. (2017). "AN IMAGE RECONSTRUCTION FRAMEWORK BASED ON DEEP NEURAL NETWORK FOR ELECTRICAL IMPEDANCE TOMOGRAPHY." Web.
1. Xiuyan Li, Yang Lu,Jianming Wang,Xin Dang,Qi Wang,Xiaojie Duan,Yukuan Sun. AN IMAGE RECONSTRUCTION FRAMEWORK BASED ON DEEP NEURAL NETWORK FOR ELECTRICAL IMPEDANCE TOMOGRAPHY [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2078

FAST DE-STREAKING METHOD USING PLAIN NEURAL NETWORK

Paper Details

Authors:
Yuxiang Li, Bo Zhang, Raoul Florent
Submitted On:
14 September 2017 - 6:40am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

presentation_icip2017b.pdf

(6 downloads)

Keywords

Subscribe

[1] Yuxiang Li, Bo Zhang, Raoul Florent, "FAST DE-STREAKING METHOD USING PLAIN NEURAL NETWORK", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2010. Accessed: Sep. 20, 2017.
@article{2010-17,
url = {http://sigport.org/2010},
author = {Yuxiang Li; Bo Zhang; Raoul Florent },
publisher = {IEEE SigPort},
title = {FAST DE-STREAKING METHOD USING PLAIN NEURAL NETWORK},
year = {2017} }
TY - EJOUR
T1 - FAST DE-STREAKING METHOD USING PLAIN NEURAL NETWORK
AU - Yuxiang Li; Bo Zhang; Raoul Florent
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2010
ER -
Yuxiang Li, Bo Zhang, Raoul Florent. (2017). FAST DE-STREAKING METHOD USING PLAIN NEURAL NETWORK. IEEE SigPort. http://sigport.org/2010
Yuxiang Li, Bo Zhang, Raoul Florent, 2017. FAST DE-STREAKING METHOD USING PLAIN NEURAL NETWORK. Available at: http://sigport.org/2010.
Yuxiang Li, Bo Zhang, Raoul Florent. (2017). "FAST DE-STREAKING METHOD USING PLAIN NEURAL NETWORK." Web.
1. Yuxiang Li, Bo Zhang, Raoul Florent. FAST DE-STREAKING METHOD USING PLAIN NEURAL NETWORK [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2010

TAD16K: An Enhanced Benchmark for Autonomous Driving


Although promising results have been achieved in the areas of object detection and classification, few works have provided an end-to-end solution to the perception problems in the autonomous driving field. In this paper, we make two contributions. Firstly, we fully enhanced our previously released TT100K benchmark and provide 16,817 elaborately labeled Tencent Street View panoramas.

Paper Details

Authors:
Yuming Li, Jue Wang, Tengfei Xing, Tianlu Liu, Chengjun Li, Kuifeng Su
Submitted On:
14 September 2017 - 6:10am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP2017_poster.pdf

(8 downloads)

Keywords

Subscribe

[1] Yuming Li, Jue Wang, Tengfei Xing, Tianlu Liu, Chengjun Li, Kuifeng Su, "TAD16K: An Enhanced Benchmark for Autonomous Driving", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2007. Accessed: Sep. 20, 2017.
@article{2007-17,
url = {http://sigport.org/2007},
author = {Yuming Li; Jue Wang; Tengfei Xing; Tianlu Liu; Chengjun Li; Kuifeng Su },
publisher = {IEEE SigPort},
title = {TAD16K: An Enhanced Benchmark for Autonomous Driving},
year = {2017} }
TY - EJOUR
T1 - TAD16K: An Enhanced Benchmark for Autonomous Driving
AU - Yuming Li; Jue Wang; Tengfei Xing; Tianlu Liu; Chengjun Li; Kuifeng Su
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2007
ER -
Yuming Li, Jue Wang, Tengfei Xing, Tianlu Liu, Chengjun Li, Kuifeng Su. (2017). TAD16K: An Enhanced Benchmark for Autonomous Driving. IEEE SigPort. http://sigport.org/2007
Yuming Li, Jue Wang, Tengfei Xing, Tianlu Liu, Chengjun Li, Kuifeng Su, 2017. TAD16K: An Enhanced Benchmark for Autonomous Driving. Available at: http://sigport.org/2007.
Yuming Li, Jue Wang, Tengfei Xing, Tianlu Liu, Chengjun Li, Kuifeng Su. (2017). "TAD16K: An Enhanced Benchmark for Autonomous Driving." Web.
1. Yuming Li, Jue Wang, Tengfei Xing, Tianlu Liu, Chengjun Li, Kuifeng Su. TAD16K: An Enhanced Benchmark for Autonomous Driving [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2007

Phase Retrieval via Coordinate Descent


Phase retrieval refers to recovery of a signal-of-interest given only the intensity measurement samples and has wide applicability including important areas of astronomy, computational biology, crystallography, digital communications, electron microscopy, neutron radiography and optical imaging. The classical problem formulation is to restore the time-domain signal from its power spectrum observations, although the Fourier transform can be generalized to any linear mappings.

Paper Details

Authors:
Submitted On:
28 June 2017 - 11:19pm
Short Link:
Type:

Document Files

phase_retrieval.pdf

(71 downloads)

Keywords

Subscribe

[1] , "Phase Retrieval via Coordinate Descent", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1799. Accessed: Sep. 20, 2017.
@article{1799-17,
url = {http://sigport.org/1799},
author = { },
publisher = {IEEE SigPort},
title = {Phase Retrieval via Coordinate Descent},
year = {2017} }
TY - EJOUR
T1 - Phase Retrieval via Coordinate Descent
AU -
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1799
ER -
. (2017). Phase Retrieval via Coordinate Descent. IEEE SigPort. http://sigport.org/1799
, 2017. Phase Retrieval via Coordinate Descent. Available at: http://sigport.org/1799.
. (2017). "Phase Retrieval via Coordinate Descent." Web.
1. . Phase Retrieval via Coordinate Descent [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1799

Robust Matrix Completion via Alternating Projection


Matrix completion aims to find the missing entries from incomplete observations using the low-rank property. Conventional convex optimization based techniques minimize the nuclear norm subject to a constraint on the Frobenius norm of the residual. However, they are not robust to outliers and have a high computational complexity. Different from the existing schemes based on solving a minimization problem, we formulate matrix completion as a feasibility problem.

Paper Details

Authors:
Submitted On:
19 June 2017 - 11:39pm
Short Link:
Type:

Document Files

AP_matrix_completion.pdf

(79 downloads)

Keywords

Subscribe

[1] , "Robust Matrix Completion via Alternating Projection", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1798. Accessed: Sep. 20, 2017.
@article{1798-17,
url = {http://sigport.org/1798},
author = { },
publisher = {IEEE SigPort},
title = {Robust Matrix Completion via Alternating Projection},
year = {2017} }
TY - EJOUR
T1 - Robust Matrix Completion via Alternating Projection
AU -
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1798
ER -
. (2017). Robust Matrix Completion via Alternating Projection. IEEE SigPort. http://sigport.org/1798
, 2017. Robust Matrix Completion via Alternating Projection. Available at: http://sigport.org/1798.
. (2017). "Robust Matrix Completion via Alternating Projection." Web.
1. . Robust Matrix Completion via Alternating Projection [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1798

AFFECT RECOGNITION FROM LIP ARTICULATIONS

Paper Details

Authors:
Submitted On:
23 March 2017 - 1:40pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

Poster_ICASSP17_Rizwan.pdf

(74 downloads)

Keywords

Subscribe

[1] , "AFFECT RECOGNITION FROM LIP ARTICULATIONS", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1781. Accessed: Sep. 20, 2017.
@article{1781-17,
url = {http://sigport.org/1781},
author = { },
publisher = {IEEE SigPort},
title = {AFFECT RECOGNITION FROM LIP ARTICULATIONS},
year = {2017} }
TY - EJOUR
T1 - AFFECT RECOGNITION FROM LIP ARTICULATIONS
AU -
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1781
ER -
. (2017). AFFECT RECOGNITION FROM LIP ARTICULATIONS. IEEE SigPort. http://sigport.org/1781
, 2017. AFFECT RECOGNITION FROM LIP ARTICULATIONS. Available at: http://sigport.org/1781.
. (2017). "AFFECT RECOGNITION FROM LIP ARTICULATIONS." Web.
1. . AFFECT RECOGNITION FROM LIP ARTICULATIONS [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1781

Disjunctive Normal Shape Boltzmann Machine


Shape Boltzmann machine (a type of Deep Boltzmann machine) is a powerful tool for shape modelling; however, has some drawbacks in representation of local shape parts. Disjunctive Normal Shape Model (DNSM) is a strong shape model that can effectively represent local parts of objects. In this paper, we propose a new shape model based on Shape Boltzmann Machine and Disjunctive Normal Shape Model which we call Disjunctive Normal Shape Boltzmann Machine (DNSBM).

Paper Details

Authors:
Ertunc Erdil, Fitsum Mesadi, Tolga Tasdizen, Mujdat Cetin
Submitted On:
13 March 2017 - 3:59pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

erdil_ICASSP17_presentation.pdf

(69 downloads)

Keywords

Subscribe

[1] Ertunc Erdil, Fitsum Mesadi, Tolga Tasdizen, Mujdat Cetin, "Disjunctive Normal Shape Boltzmann Machine", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1760. Accessed: Sep. 20, 2017.
@article{1760-17,
url = {http://sigport.org/1760},
author = {Ertunc Erdil; Fitsum Mesadi; Tolga Tasdizen; Mujdat Cetin },
publisher = {IEEE SigPort},
title = {Disjunctive Normal Shape Boltzmann Machine},
year = {2017} }
TY - EJOUR
T1 - Disjunctive Normal Shape Boltzmann Machine
AU - Ertunc Erdil; Fitsum Mesadi; Tolga Tasdizen; Mujdat Cetin
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1760
ER -
Ertunc Erdil, Fitsum Mesadi, Tolga Tasdizen, Mujdat Cetin. (2017). Disjunctive Normal Shape Boltzmann Machine. IEEE SigPort. http://sigport.org/1760
Ertunc Erdil, Fitsum Mesadi, Tolga Tasdizen, Mujdat Cetin, 2017. Disjunctive Normal Shape Boltzmann Machine. Available at: http://sigport.org/1760.
Ertunc Erdil, Fitsum Mesadi, Tolga Tasdizen, Mujdat Cetin. (2017). "Disjunctive Normal Shape Boltzmann Machine." Web.
1. Ertunc Erdil, Fitsum Mesadi, Tolga Tasdizen, Mujdat Cetin. Disjunctive Normal Shape Boltzmann Machine [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1760

A Distributed Constrained-Form Support Vector Machine


Despite the importance of distributed learning, few fully distributed support vector machines exist. In this paper, not only do we provide a fully distributed nonlinear SVM; we propose the first distributed constrained-form SVM. In the fully distributed context, a dataset is distributed among networked agents that cannot divulge their data, let alone centralize the data, and can only communicate with their neighbors in the network. Our strategy is based on two algorithms: the Douglas-Rachford algorithm and the projection-gradient method.

Poster.pdf

PDF icon Poster.pdf (163 downloads)

Paper Details

Authors:
François D. Côté, Ioannis N. Psaromiligkos, Warren J. Gross
Submitted On:
9 March 2017 - 12:26pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster.pdf

(163 downloads)

Keywords

Additional Categories

Subscribe

[1] François D. Côté, Ioannis N. Psaromiligkos, Warren J. Gross, "A Distributed Constrained-Form Support Vector Machine ", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1723. Accessed: Sep. 20, 2017.
@article{1723-17,
url = {http://sigport.org/1723},
author = {François D. Côté; Ioannis N. Psaromiligkos; Warren J. Gross },
publisher = {IEEE SigPort},
title = {A Distributed Constrained-Form Support Vector Machine },
year = {2017} }
TY - EJOUR
T1 - A Distributed Constrained-Form Support Vector Machine
AU - François D. Côté; Ioannis N. Psaromiligkos; Warren J. Gross
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1723
ER -
François D. Côté, Ioannis N. Psaromiligkos, Warren J. Gross. (2017). A Distributed Constrained-Form Support Vector Machine . IEEE SigPort. http://sigport.org/1723
François D. Côté, Ioannis N. Psaromiligkos, Warren J. Gross, 2017. A Distributed Constrained-Form Support Vector Machine . Available at: http://sigport.org/1723.
François D. Côté, Ioannis N. Psaromiligkos, Warren J. Gross. (2017). "A Distributed Constrained-Form Support Vector Machine ." Web.
1. François D. Côté, Ioannis N. Psaromiligkos, Warren J. Gross. A Distributed Constrained-Form Support Vector Machine [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1723

Pages