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

ICASSP 2019

ICASSP is the world’s largest and most comprehensive technical conference focused on signal processing and its applications. The 2019 conference will feature world-class presentations by internationally renowned speakers, cutting-edge session topics and provide a fantastic opportunity to network with like-minded professionals from around the world. Visit website

Inferring Private Information in Wireless Sensor Networks

Paper Details

Authors:
Submitted On:
7 May 2019 - 5:51pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP_paper3565.pdf

(55)

Subscribe

[1] , "Inferring Private Information in Wireless Sensor Networks", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3963. Accessed: Apr. 04, 2020.
@article{3963-19,
url = {http://sigport.org/3963},
author = { },
publisher = {IEEE SigPort},
title = {Inferring Private Information in Wireless Sensor Networks},
year = {2019} }
TY - EJOUR
T1 - Inferring Private Information in Wireless Sensor Networks
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3963
ER -
. (2019). Inferring Private Information in Wireless Sensor Networks. IEEE SigPort. http://sigport.org/3963
, 2019. Inferring Private Information in Wireless Sensor Networks. Available at: http://sigport.org/3963.
. (2019). "Inferring Private Information in Wireless Sensor Networks." Web.
1. . Inferring Private Information in Wireless Sensor Networks [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3963

DISTRIBUTED TRACKING OF MANEUVERING TARGET: A FINITE-TIME ALGORITHM

Paper Details

Authors:
Submitted On:
7 May 2019 - 5:46pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

poster

(48)

Subscribe

[1] , "DISTRIBUTED TRACKING OF MANEUVERING TARGET: A FINITE-TIME ALGORITHM", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3962. Accessed: Apr. 04, 2020.
@article{3962-19,
url = {http://sigport.org/3962},
author = { },
publisher = {IEEE SigPort},
title = {DISTRIBUTED TRACKING OF MANEUVERING TARGET: A FINITE-TIME ALGORITHM},
year = {2019} }
TY - EJOUR
T1 - DISTRIBUTED TRACKING OF MANEUVERING TARGET: A FINITE-TIME ALGORITHM
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3962
ER -
. (2019). DISTRIBUTED TRACKING OF MANEUVERING TARGET: A FINITE-TIME ALGORITHM. IEEE SigPort. http://sigport.org/3962
, 2019. DISTRIBUTED TRACKING OF MANEUVERING TARGET: A FINITE-TIME ALGORITHM. Available at: http://sigport.org/3962.
. (2019). "DISTRIBUTED TRACKING OF MANEUVERING TARGET: A FINITE-TIME ALGORITHM." Web.
1. . DISTRIBUTED TRACKING OF MANEUVERING TARGET: A FINITE-TIME ALGORITHM [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3962

DEEP TEMPORAL PYRAMID DESIGN FOR ACTION RECOGNITION


Deep convolutional neural networks (CNNs) are nowadays achieving significant leaps in different pattern recognition tasks including action recognition. Current CNNs are increasingly deeper, data-hungrier and this makes their success tributary of the abundance of labeled training data. CNNs also rely on max/average pooling which reduces dimensionality of output layers and hence attenuates their sensitivity to the availability of labeled data.

Paper Details

Authors:
Ahmed Mazari, Hichem Sahbi
Submitted On:
7 May 2019 - 5:45pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

Poster_ICASSP19_Action_Recognition.pdf

(71)

Subscribe

[1] Ahmed Mazari, Hichem Sahbi, "DEEP TEMPORAL PYRAMID DESIGN FOR ACTION RECOGNITION", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3961. Accessed: Apr. 04, 2020.
@article{3961-19,
url = {http://sigport.org/3961},
author = {Ahmed Mazari; Hichem Sahbi },
publisher = {IEEE SigPort},
title = {DEEP TEMPORAL PYRAMID DESIGN FOR ACTION RECOGNITION},
year = {2019} }
TY - EJOUR
T1 - DEEP TEMPORAL PYRAMID DESIGN FOR ACTION RECOGNITION
AU - Ahmed Mazari; Hichem Sahbi
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3961
ER -
Ahmed Mazari, Hichem Sahbi. (2019). DEEP TEMPORAL PYRAMID DESIGN FOR ACTION RECOGNITION. IEEE SigPort. http://sigport.org/3961
Ahmed Mazari, Hichem Sahbi, 2019. DEEP TEMPORAL PYRAMID DESIGN FOR ACTION RECOGNITION. Available at: http://sigport.org/3961.
Ahmed Mazari, Hichem Sahbi. (2019). "DEEP TEMPORAL PYRAMID DESIGN FOR ACTION RECOGNITION." Web.
1. Ahmed Mazari, Hichem Sahbi. DEEP TEMPORAL PYRAMID DESIGN FOR ACTION RECOGNITION [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3961

Uplink Multi-user MIMO Detection via Parallel Access

Paper Details

Authors:
Submitted On:
7 May 2019 - 5:23pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:

Document Files

UL-MUMIMO-ICASSP-Keke Zu

(68)

Subscribe

[1] , "Uplink Multi-user MIMO Detection via Parallel Access", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3960. Accessed: Apr. 04, 2020.
@article{3960-19,
url = {http://sigport.org/3960},
author = { },
publisher = {IEEE SigPort},
title = {Uplink Multi-user MIMO Detection via Parallel Access},
year = {2019} }
TY - EJOUR
T1 - Uplink Multi-user MIMO Detection via Parallel Access
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3960
ER -
. (2019). Uplink Multi-user MIMO Detection via Parallel Access. IEEE SigPort. http://sigport.org/3960
, 2019. Uplink Multi-user MIMO Detection via Parallel Access. Available at: http://sigport.org/3960.
. (2019). "Uplink Multi-user MIMO Detection via Parallel Access." Web.
1. . Uplink Multi-user MIMO Detection via Parallel Access [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3960

CALVI_ICASSP

Paper Details

Authors:
Giuseppe G. Calvi, Vladimir Lucic, Danilo P. Mandic
Submitted On:
7 May 2019 - 5:22pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

CALVI_ICASSP.pdf

(60)

Subscribe

[1] Giuseppe G. Calvi, Vladimir Lucic, Danilo P. Mandic, "CALVI_ICASSP", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3959. Accessed: Apr. 04, 2020.
@article{3959-19,
url = {http://sigport.org/3959},
author = {Giuseppe G. Calvi; Vladimir Lucic; Danilo P. Mandic },
publisher = {IEEE SigPort},
title = {CALVI_ICASSP},
year = {2019} }
TY - EJOUR
T1 - CALVI_ICASSP
AU - Giuseppe G. Calvi; Vladimir Lucic; Danilo P. Mandic
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3959
ER -
Giuseppe G. Calvi, Vladimir Lucic, Danilo P. Mandic. (2019). CALVI_ICASSP. IEEE SigPort. http://sigport.org/3959
Giuseppe G. Calvi, Vladimir Lucic, Danilo P. Mandic, 2019. CALVI_ICASSP. Available at: http://sigport.org/3959.
Giuseppe G. Calvi, Vladimir Lucic, Danilo P. Mandic. (2019). "CALVI_ICASSP." Web.
1. Giuseppe G. Calvi, Vladimir Lucic, Danilo P. Mandic. CALVI_ICASSP [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3959

Learning to Fuse Latent Representations for Multimodal Data

Paper Details

Authors:
Oyebade Oyedotun, Djamila Aouada, Bjorn Ottersten
Submitted On:
7 May 2019 - 5:21pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

2019 SnT Poster Template_V07.pdf

(48)

Subscribe

[1] Oyebade Oyedotun, Djamila Aouada, Bjorn Ottersten, "Learning to Fuse Latent Representations for Multimodal Data", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3958. Accessed: Apr. 04, 2020.
@article{3958-19,
url = {http://sigport.org/3958},
author = {Oyebade Oyedotun; Djamila Aouada; Bjorn Ottersten },
publisher = {IEEE SigPort},
title = {Learning to Fuse Latent Representations for Multimodal Data},
year = {2019} }
TY - EJOUR
T1 - Learning to Fuse Latent Representations for Multimodal Data
AU - Oyebade Oyedotun; Djamila Aouada; Bjorn Ottersten
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3958
ER -
Oyebade Oyedotun, Djamila Aouada, Bjorn Ottersten. (2019). Learning to Fuse Latent Representations for Multimodal Data. IEEE SigPort. http://sigport.org/3958
Oyebade Oyedotun, Djamila Aouada, Bjorn Ottersten, 2019. Learning to Fuse Latent Representations for Multimodal Data. Available at: http://sigport.org/3958.
Oyebade Oyedotun, Djamila Aouada, Bjorn Ottersten. (2019). "Learning to Fuse Latent Representations for Multimodal Data." Web.
1. Oyebade Oyedotun, Djamila Aouada, Bjorn Ottersten. Learning to Fuse Latent Representations for Multimodal Data [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3958

Real-time tracker with fast recovery from target loss


In this paper, we introduce a variation of a state-of-the-art real-time tracker (CFNet), which adds to the original algorithm robustness to target loss without a significant computational overhead. The new method is based on the assumption that the feature map can be used to estimate the tracking confidence more accurately.

Paper Details

Authors:
Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli
Submitted On:
7 May 2019 - 5:10pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:

Document Files

FRTL_poster.pdf

(38)

Subscribe

[1] Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli, "Real-time tracker with fast recovery from target loss", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3957. Accessed: Apr. 04, 2020.
@article{3957-19,
url = {http://sigport.org/3957},
author = {Alessandro Bay; Panagiotis Sidiropoulos; Eduard Vazquez; Michele Sasdelli },
publisher = {IEEE SigPort},
title = {Real-time tracker with fast recovery from target loss},
year = {2019} }
TY - EJOUR
T1 - Real-time tracker with fast recovery from target loss
AU - Alessandro Bay; Panagiotis Sidiropoulos; Eduard Vazquez; Michele Sasdelli
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3957
ER -
Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli. (2019). Real-time tracker with fast recovery from target loss. IEEE SigPort. http://sigport.org/3957
Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli, 2019. Real-time tracker with fast recovery from target loss. Available at: http://sigport.org/3957.
Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli. (2019). "Real-time tracker with fast recovery from target loss." Web.
1. Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli. Real-time tracker with fast recovery from target loss [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3957

A Non-Convex Approach to Non-negative Super-Resolution: Theory and Algorithm

Paper Details

Authors:
Submitted On:
7 May 2019 - 5:09pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

QIAO_HENG.pdf

(80)

Subscribe

[1] , "A Non-Convex Approach to Non-negative Super-Resolution: Theory and Algorithm", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3956. Accessed: Apr. 04, 2020.
@article{3956-19,
url = {http://sigport.org/3956},
author = { },
publisher = {IEEE SigPort},
title = {A Non-Convex Approach to Non-negative Super-Resolution: Theory and Algorithm},
year = {2019} }
TY - EJOUR
T1 - A Non-Convex Approach to Non-negative Super-Resolution: Theory and Algorithm
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3956
ER -
. (2019). A Non-Convex Approach to Non-negative Super-Resolution: Theory and Algorithm. IEEE SigPort. http://sigport.org/3956
, 2019. A Non-Convex Approach to Non-negative Super-Resolution: Theory and Algorithm. Available at: http://sigport.org/3956.
. (2019). "A Non-Convex Approach to Non-negative Super-Resolution: Theory and Algorithm." Web.
1. . A Non-Convex Approach to Non-negative Super-Resolution: Theory and Algorithm [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3956

Unsuper vised Deep Clustering for Source Separation: Direct Learning from Mixtures Using Spatial Information Slides


We present a monophonic source separation system that is trained by only observing mixtures with no ground truth separation information. We use a deep clustering approach which trains on multi-channel mixtures and learns to project spectrogram bins to source clusters that correlate with various spatial features. We show that using such a training process we can obtain separation performance that is as good as making use of ground truth separation information.

Paper Details

Authors:
Efthymios Tzinis, Shrikant Venkataramani, Paris Smaragdis
Submitted On:
7 May 2019 - 4:28pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

unsup_dc_icassp19_slides

(66)

Subscribe

[1] Efthymios Tzinis, Shrikant Venkataramani, Paris Smaragdis, "Unsuper vised Deep Clustering for Source Separation: Direct Learning from Mixtures Using Spatial Information Slides", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3954. Accessed: Apr. 04, 2020.
@article{3954-19,
url = {http://sigport.org/3954},
author = {Efthymios Tzinis; Shrikant Venkataramani; Paris Smaragdis },
publisher = {IEEE SigPort},
title = {Unsuper vised Deep Clustering for Source Separation: Direct Learning from Mixtures Using Spatial Information Slides},
year = {2019} }
TY - EJOUR
T1 - Unsuper vised Deep Clustering for Source Separation: Direct Learning from Mixtures Using Spatial Information Slides
AU - Efthymios Tzinis; Shrikant Venkataramani; Paris Smaragdis
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3954
ER -
Efthymios Tzinis, Shrikant Venkataramani, Paris Smaragdis. (2019). Unsuper vised Deep Clustering for Source Separation: Direct Learning from Mixtures Using Spatial Information Slides. IEEE SigPort. http://sigport.org/3954
Efthymios Tzinis, Shrikant Venkataramani, Paris Smaragdis, 2019. Unsuper vised Deep Clustering for Source Separation: Direct Learning from Mixtures Using Spatial Information Slides. Available at: http://sigport.org/3954.
Efthymios Tzinis, Shrikant Venkataramani, Paris Smaragdis. (2019). "Unsuper vised Deep Clustering for Source Separation: Direct Learning from Mixtures Using Spatial Information Slides." Web.
1. Efthymios Tzinis, Shrikant Venkataramani, Paris Smaragdis. Unsuper vised Deep Clustering for Source Separation: Direct Learning from Mixtures Using Spatial Information Slides [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3954

Word Characters and Phone Pronunciation Embedding for ASR Confidence Classifier


Confidences are integral to ASR systems, and applied to data selection, adaptation, ranking hypotheses, arbitration etc.Hybrid ASR system is inherently a match between pronunciations and AM+LM evidence but current confidence features lack pronunciation information. We develop pronunciation embeddings to represent and factorize acoustic score in relevant bases, and demonstrate 8-10% relative reduction in false alarm (FA) on large scale tasks. We generalize to standard NLP embeddings like Glove, and show 16% relative reduction in FA in combination with Glove.

Paper Details

Authors:
Kshitiz Kumar, Tasos Anastasakos, Yifan Gong
Submitted On:
7 May 2019 - 3:33pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

WordEmbed_v5.pdf

(81)

Subscribe

[1] Kshitiz Kumar, Tasos Anastasakos, Yifan Gong, "Word Characters and Phone Pronunciation Embedding for ASR Confidence Classifier", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3953. Accessed: Apr. 04, 2020.
@article{3953-19,
url = {http://sigport.org/3953},
author = {Kshitiz Kumar; Tasos Anastasakos; Yifan Gong },
publisher = {IEEE SigPort},
title = {Word Characters and Phone Pronunciation Embedding for ASR Confidence Classifier},
year = {2019} }
TY - EJOUR
T1 - Word Characters and Phone Pronunciation Embedding for ASR Confidence Classifier
AU - Kshitiz Kumar; Tasos Anastasakos; Yifan Gong
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3953
ER -
Kshitiz Kumar, Tasos Anastasakos, Yifan Gong. (2019). Word Characters and Phone Pronunciation Embedding for ASR Confidence Classifier. IEEE SigPort. http://sigport.org/3953
Kshitiz Kumar, Tasos Anastasakos, Yifan Gong, 2019. Word Characters and Phone Pronunciation Embedding for ASR Confidence Classifier. Available at: http://sigport.org/3953.
Kshitiz Kumar, Tasos Anastasakos, Yifan Gong. (2019). "Word Characters and Phone Pronunciation Embedding for ASR Confidence Classifier." Web.
1. Kshitiz Kumar, Tasos Anastasakos, Yifan Gong. Word Characters and Phone Pronunciation Embedding for ASR Confidence Classifier [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3953

Pages