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

Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks


Predicting the gaze of a user can have important applications in hu- man computer interactions (HCI). They find applications in areas such as social interaction, driver distraction, human robot interaction and education. Appearance based models for gaze estimation have significantly improved due to recent advances in convolutional neural network (CNN). This paper proposes a method to predict the gaze of a user with deep models purely based on CNNs.

Paper Details

Authors:
Sumit Jha, Carlos Busso
Submitted On:
20 May 2020 - 9:53am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

Jha_2019-poster.pdf

(4)

Subscribe

[1] Sumit Jha, Carlos Busso, "Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5410. Accessed: May. 30, 2020.
@article{5410-20,
url = {http://sigport.org/5410},
author = {Sumit Jha; Carlos Busso },
publisher = {IEEE SigPort},
title = {Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks},
year = {2020} }
TY - EJOUR
T1 - Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks
AU - Sumit Jha; Carlos Busso
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5410
ER -
Sumit Jha, Carlos Busso. (2020). Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks. IEEE SigPort. http://sigport.org/5410
Sumit Jha, Carlos Busso, 2020. Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks. Available at: http://sigport.org/5410.
Sumit Jha, Carlos Busso. (2020). "Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks." Web.
1. Sumit Jha, Carlos Busso. Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5410

Retrieving speech samples with similar emotional content using a triplet loss function


The ability to identify speech with similar emotional content is valuable to many applications, including speech retrieval, surveil- lance, and emotional speech synthesis. While current formulations in speech emotion recognition based on classification or regression are not appropriate for this task, solutions based on preference learn- ing offer appealing approaches for this task. This paper aims to find speech samples that are emotionally similar to an anchor speech sample provided as a query. This novel formulation opens interest- ing research questions.

Paper Details

Authors:
John Harvill, Mohammed AbdelWahab, Reza Lotfian, Carlos Busso
Submitted On:
20 May 2020 - 9:50am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

Harvill_2019-poster.pdf

(6)

Subscribe

[1] John Harvill, Mohammed AbdelWahab, Reza Lotfian, Carlos Busso, "Retrieving speech samples with similar emotional content using a triplet loss function", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5409. Accessed: May. 30, 2020.
@article{5409-20,
url = {http://sigport.org/5409},
author = {John Harvill; Mohammed AbdelWahab; Reza Lotfian; Carlos Busso },
publisher = {IEEE SigPort},
title = {Retrieving speech samples with similar emotional content using a triplet loss function},
year = {2020} }
TY - EJOUR
T1 - Retrieving speech samples with similar emotional content using a triplet loss function
AU - John Harvill; Mohammed AbdelWahab; Reza Lotfian; Carlos Busso
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5409
ER -
John Harvill, Mohammed AbdelWahab, Reza Lotfian, Carlos Busso. (2020). Retrieving speech samples with similar emotional content using a triplet loss function. IEEE SigPort. http://sigport.org/5409
John Harvill, Mohammed AbdelWahab, Reza Lotfian, Carlos Busso, 2020. Retrieving speech samples with similar emotional content using a triplet loss function. Available at: http://sigport.org/5409.
John Harvill, Mohammed AbdelWahab, Reza Lotfian, Carlos Busso. (2020). "Retrieving speech samples with similar emotional content using a triplet loss function." Web.
1. John Harvill, Mohammed AbdelWahab, Reza Lotfian, Carlos Busso. Retrieving speech samples with similar emotional content using a triplet loss function [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5409

On the Transferability of Adversarial Examples Against CNN-Based Image Forensics


Recent studies have shown that Convolutional Neural Networks (CNN) are relatively easy to attack through the generation of so-called adversarial examples. Such vulnerability also affects CNN-based image forensic tools. Research in deep learning has shown that adversarial examples exhibit a certain degree of transferability, i.e., they maintain part of their effectiveness even against CNN models other than the one targeted by the attack. This is a very strong property undermining the usability of CNN’s in security-oriented applications.

Paper Details

Authors:
Mauro Barni, Kassem Kallas, Ehsan Nowroozi, Benedetta Tondi
Submitted On:
30 January 2020 - 12:18pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

ICASSP 2019.pdf

(66)

Subscribe

[1] Mauro Barni, Kassem Kallas, Ehsan Nowroozi, Benedetta Tondi, "On the Transferability of Adversarial Examples Against CNN-Based Image Forensics", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/4969. Accessed: May. 30, 2020.
@article{4969-20,
url = {http://sigport.org/4969},
author = {Mauro Barni; Kassem Kallas; Ehsan Nowroozi; Benedetta Tondi },
publisher = {IEEE SigPort},
title = {On the Transferability of Adversarial Examples Against CNN-Based Image Forensics},
year = {2020} }
TY - EJOUR
T1 - On the Transferability of Adversarial Examples Against CNN-Based Image Forensics
AU - Mauro Barni; Kassem Kallas; Ehsan Nowroozi; Benedetta Tondi
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/4969
ER -
Mauro Barni, Kassem Kallas, Ehsan Nowroozi, Benedetta Tondi. (2020). On the Transferability of Adversarial Examples Against CNN-Based Image Forensics. IEEE SigPort. http://sigport.org/4969
Mauro Barni, Kassem Kallas, Ehsan Nowroozi, Benedetta Tondi, 2020. On the Transferability of Adversarial Examples Against CNN-Based Image Forensics. Available at: http://sigport.org/4969.
Mauro Barni, Kassem Kallas, Ehsan Nowroozi, Benedetta Tondi. (2020). "On the Transferability of Adversarial Examples Against CNN-Based Image Forensics." Web.
1. Mauro Barni, Kassem Kallas, Ehsan Nowroozi, Benedetta Tondi. On the Transferability of Adversarial Examples Against CNN-Based Image Forensics [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/4969

LMS: PAST, PRESENT AND FUTURE: Puzzles, Problems and Potentials


We give a brief history of the performance analysis of LMS.
Using averaging theory we show when and why the ‘independence
assumption’ ‘works’; we preface this with a fast
heuristic explanation of averaging methods, clarifying their
connection to the ‘ODE’ method. We then extend the discussion
to more recent distributed versions such as diffusion
LMS and consensus. While single node LMS is a single timescale
algorithm it turns out that distributed versions are twotime
scale systems, something that is not yet widely understood.

Paper Details

Authors:
Submitted On:
25 June 2019 - 12:16am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

vsLMS-ICASSP19.pdf

(156)

Subscribe

[1] , "LMS: PAST, PRESENT AND FUTURE: Puzzles, Problems and Potentials", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4566. Accessed: May. 30, 2020.
@article{4566-19,
url = {http://sigport.org/4566},
author = { },
publisher = {IEEE SigPort},
title = {LMS: PAST, PRESENT AND FUTURE: Puzzles, Problems and Potentials},
year = {2019} }
TY - EJOUR
T1 - LMS: PAST, PRESENT AND FUTURE: Puzzles, Problems and Potentials
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4566
ER -
. (2019). LMS: PAST, PRESENT AND FUTURE: Puzzles, Problems and Potentials. IEEE SigPort. http://sigport.org/4566
, 2019. LMS: PAST, PRESENT AND FUTURE: Puzzles, Problems and Potentials. Available at: http://sigport.org/4566.
. (2019). "LMS: PAST, PRESENT AND FUTURE: Puzzles, Problems and Potentials." Web.
1. . LMS: PAST, PRESENT AND FUTURE: Puzzles, Problems and Potentials [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4566

Speech Landmark Bigrams for Depression Detection from Naturalistic Smartphone Speech


Detection of depression from speech has attracted significant research attention in recent years but remains a challenge, particularly for speech from diverse smartphones in natural environments. This paper proposes two sets of novel features based on speech landmark bigrams associated with abrupt speech articulatory events for depression detection from smartphone audio recordings. Combined with techniques adapted from natural language text processing, the proposed features further exploit landmark bigrams by discovering latent articulatory events.

Paper Details

Authors:
Zhaocheng Huang, Julien Epps, Dale Joachim
Submitted On:
6 June 2019 - 4:42am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP2019_Huang_V01_uploaded.pdf

(164)

Subscribe

[1] Zhaocheng Huang, Julien Epps, Dale Joachim, "Speech Landmark Bigrams for Depression Detection from Naturalistic Smartphone Speech", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4565. Accessed: May. 30, 2020.
@article{4565-19,
url = {http://sigport.org/4565},
author = {Zhaocheng Huang; Julien Epps; Dale Joachim },
publisher = {IEEE SigPort},
title = {Speech Landmark Bigrams for Depression Detection from Naturalistic Smartphone Speech},
year = {2019} }
TY - EJOUR
T1 - Speech Landmark Bigrams for Depression Detection from Naturalistic Smartphone Speech
AU - Zhaocheng Huang; Julien Epps; Dale Joachim
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4565
ER -
Zhaocheng Huang, Julien Epps, Dale Joachim. (2019). Speech Landmark Bigrams for Depression Detection from Naturalistic Smartphone Speech. IEEE SigPort. http://sigport.org/4565
Zhaocheng Huang, Julien Epps, Dale Joachim, 2019. Speech Landmark Bigrams for Depression Detection from Naturalistic Smartphone Speech. Available at: http://sigport.org/4565.
Zhaocheng Huang, Julien Epps, Dale Joachim. (2019). "Speech Landmark Bigrams for Depression Detection from Naturalistic Smartphone Speech." Web.
1. Zhaocheng Huang, Julien Epps, Dale Joachim. Speech Landmark Bigrams for Depression Detection from Naturalistic Smartphone Speech [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4565

Graph Signal Sampling via Reinforcement Learning


We model the sampling and recovery of clustered graph signals as a reinforcement learning (RL) problem. The signal sampling is carried out by an agent which crawls over the graph and selects the most relevant graph nodes to sample. The goal of the agent is to select signal samples which allow for the most accurate recovery. The sample selection is formulated as a multi-armed bandit (MAB) problem, which lends naturally to learning efficient sampling strategies using the well-known gradient MAB algorithm.

Paper Details

Authors:
Oleksii Abramenko, Alexander Jung
Submitted On:
30 May 2019 - 10:50am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster_Abramenko_Jung.pdf

(131)

Keywords

Additional Categories

Subscribe

[1] Oleksii Abramenko, Alexander Jung, "Graph Signal Sampling via Reinforcement Learning", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4562. Accessed: May. 30, 2020.
@article{4562-19,
url = {http://sigport.org/4562},
author = {Oleksii Abramenko; Alexander Jung },
publisher = {IEEE SigPort},
title = {Graph Signal Sampling via Reinforcement Learning},
year = {2019} }
TY - EJOUR
T1 - Graph Signal Sampling via Reinforcement Learning
AU - Oleksii Abramenko; Alexander Jung
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4562
ER -
Oleksii Abramenko, Alexander Jung. (2019). Graph Signal Sampling via Reinforcement Learning. IEEE SigPort. http://sigport.org/4562
Oleksii Abramenko, Alexander Jung, 2019. Graph Signal Sampling via Reinforcement Learning. Available at: http://sigport.org/4562.
Oleksii Abramenko, Alexander Jung. (2019). "Graph Signal Sampling via Reinforcement Learning." Web.
1. Oleksii Abramenko, Alexander Jung. Graph Signal Sampling via Reinforcement Learning [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4562

ROBUST M-ESTIMATION BASED MATRIX COMPLETION


Conventional approaches to matrix completion are sensitive to outliers and impulsive noise. This paper develops robust and computationally efficient M-estimation based matrix completion algorithms. By appropriately arranging the observed entries, and then applying alternating minimization, the robust matrix completion problem is converted into a set of regression M-estimation problems. Making use of differ- entiable loss functions, the proposed algorithm overcomes a weakness of the lp-loss (p ≤ 1), which easily gets stuck in an inferior point.

Paper Details

Authors:
Michael Muma, Wen-Jun Zeng, Abdelhak M. Zoubir
Submitted On:
27 May 2019 - 11:28am
Short Link:
Type:
Event:
Paper Code:
Document Year:
Cite

Document Files

ICASSP_2019_Robust_M_Estimation_Based_Matrix_Completion_Poster.pdf

(126)

Subscribe

[1] Michael Muma, Wen-Jun Zeng, Abdelhak M. Zoubir, "ROBUST M-ESTIMATION BASED MATRIX COMPLETION", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4561. Accessed: May. 30, 2020.
@article{4561-19,
url = {http://sigport.org/4561},
author = {Michael Muma; Wen-Jun Zeng; Abdelhak M. Zoubir },
publisher = {IEEE SigPort},
title = {ROBUST M-ESTIMATION BASED MATRIX COMPLETION},
year = {2019} }
TY - EJOUR
T1 - ROBUST M-ESTIMATION BASED MATRIX COMPLETION
AU - Michael Muma; Wen-Jun Zeng; Abdelhak M. Zoubir
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4561
ER -
Michael Muma, Wen-Jun Zeng, Abdelhak M. Zoubir. (2019). ROBUST M-ESTIMATION BASED MATRIX COMPLETION. IEEE SigPort. http://sigport.org/4561
Michael Muma, Wen-Jun Zeng, Abdelhak M. Zoubir, 2019. ROBUST M-ESTIMATION BASED MATRIX COMPLETION. Available at: http://sigport.org/4561.
Michael Muma, Wen-Jun Zeng, Abdelhak M. Zoubir. (2019). "ROBUST M-ESTIMATION BASED MATRIX COMPLETION." Web.
1. Michael Muma, Wen-Jun Zeng, Abdelhak M. Zoubir. ROBUST M-ESTIMATION BASED MATRIX COMPLETION [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4561

When can a System of Subnetworks be Registered Uniquely?


Consider a network with N nodes in d dimensions, and M overlapping subsets P_1,...,P_M (subnetworks). Assume that the nodes in a given P_i are observed in a local coordinate system. We wish to register the subnetworks using the knowledge of the observed coordinates. More precisely, we want to compute the positions of the N nodes in a global coordinate system, given P_1,...,P_M and the corresponding local coordinates. Among other applications, this problem arises in divide-and-conquer algorithms for localization of adhoc sensor networks.

Paper Details

Authors:
Aditya V. Singh, Kunal N. Chaudhury
Submitted On:
27 May 2019 - 5:42am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Unique Point Cloud Registration

(101)

Subscribe

[1] Aditya V. Singh, Kunal N. Chaudhury, "When can a System of Subnetworks be Registered Uniquely?", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4560. Accessed: May. 30, 2020.
@article{4560-19,
url = {http://sigport.org/4560},
author = {Aditya V. Singh; Kunal N. Chaudhury },
publisher = {IEEE SigPort},
title = {When can a System of Subnetworks be Registered Uniquely?},
year = {2019} }
TY - EJOUR
T1 - When can a System of Subnetworks be Registered Uniquely?
AU - Aditya V. Singh; Kunal N. Chaudhury
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4560
ER -
Aditya V. Singh, Kunal N. Chaudhury. (2019). When can a System of Subnetworks be Registered Uniquely?. IEEE SigPort. http://sigport.org/4560
Aditya V. Singh, Kunal N. Chaudhury, 2019. When can a System of Subnetworks be Registered Uniquely?. Available at: http://sigport.org/4560.
Aditya V. Singh, Kunal N. Chaudhury. (2019). "When can a System of Subnetworks be Registered Uniquely?." Web.
1. Aditya V. Singh, Kunal N. Chaudhury. When can a System of Subnetworks be Registered Uniquely? [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4560

Speech Emotion Recognition Using Multi-hop Attention Mechanism


In this paper, we are interested in exploiting textual and acoustic data of an utterance for the speech emotion classification task. The baseline approach models the information from audio and text independently using two deep neural networks (DNNs). The outputs from both the DNNs are then fused for classification. As opposed to using knowledge from both the modalities separately, we propose a framework to exploit acoustic information in tandem with lexical data.

Paper Details

Authors:
Seunghyun Yoon,Seokhyun Byun,Subhadeep Dey,Kyomin Jung
Submitted On:
23 May 2019 - 5:27am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

presentation slide

(229)

Subscribe

[1] Seunghyun Yoon,Seokhyun Byun,Subhadeep Dey,Kyomin Jung, "Speech Emotion Recognition Using Multi-hop Attention Mechanism", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4559. Accessed: May. 30, 2020.
@article{4559-19,
url = {http://sigport.org/4559},
author = {Seunghyun Yoon;Seokhyun Byun;Subhadeep Dey;Kyomin Jung },
publisher = {IEEE SigPort},
title = {Speech Emotion Recognition Using Multi-hop Attention Mechanism},
year = {2019} }
TY - EJOUR
T1 - Speech Emotion Recognition Using Multi-hop Attention Mechanism
AU - Seunghyun Yoon;Seokhyun Byun;Subhadeep Dey;Kyomin Jung
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4559
ER -
Seunghyun Yoon,Seokhyun Byun,Subhadeep Dey,Kyomin Jung. (2019). Speech Emotion Recognition Using Multi-hop Attention Mechanism. IEEE SigPort. http://sigport.org/4559
Seunghyun Yoon,Seokhyun Byun,Subhadeep Dey,Kyomin Jung, 2019. Speech Emotion Recognition Using Multi-hop Attention Mechanism. Available at: http://sigport.org/4559.
Seunghyun Yoon,Seokhyun Byun,Subhadeep Dey,Kyomin Jung. (2019). "Speech Emotion Recognition Using Multi-hop Attention Mechanism." Web.
1. Seunghyun Yoon,Seokhyun Byun,Subhadeep Dey,Kyomin Jung. Speech Emotion Recognition Using Multi-hop Attention Mechanism [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4559

Poster Imperceptible Audio Communication


A differential acoustic OFDM technique is presented to embed data imperceptibly in existing music. The method allows playing back music containing the data with a speaker without users noticing the embedded data channel. Using a microphone, the data can be recovered from the recording. Experiments with smartphone microphones show that transmission distances of 24 meters are possible, while achieving bit error ratios of less than 10 percent, depending on the environment.

Paper Details

Authors:
Manuel Eichelberger, Simon Tanner, Gabriel Voirol, Roger Wattenhofer
Submitted On:
22 May 2019 - 8:47am
Short Link:
Type:
Event:
Paper Code:
Document Year:
Cite

Document Files

Poster ICASSP 2019.pdf

(113)

Subscribe

[1] Manuel Eichelberger, Simon Tanner, Gabriel Voirol, Roger Wattenhofer, "Poster Imperceptible Audio Communication", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4557. Accessed: May. 30, 2020.
@article{4557-19,
url = {http://sigport.org/4557},
author = {Manuel Eichelberger; Simon Tanner; Gabriel Voirol; Roger Wattenhofer },
publisher = {IEEE SigPort},
title = {Poster Imperceptible Audio Communication},
year = {2019} }
TY - EJOUR
T1 - Poster Imperceptible Audio Communication
AU - Manuel Eichelberger; Simon Tanner; Gabriel Voirol; Roger Wattenhofer
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4557
ER -
Manuel Eichelberger, Simon Tanner, Gabriel Voirol, Roger Wattenhofer. (2019). Poster Imperceptible Audio Communication. IEEE SigPort. http://sigport.org/4557
Manuel Eichelberger, Simon Tanner, Gabriel Voirol, Roger Wattenhofer, 2019. Poster Imperceptible Audio Communication. Available at: http://sigport.org/4557.
Manuel Eichelberger, Simon Tanner, Gabriel Voirol, Roger Wattenhofer. (2019). "Poster Imperceptible Audio Communication." Web.
1. Manuel Eichelberger, Simon Tanner, Gabriel Voirol, Roger Wattenhofer. Poster Imperceptible Audio Communication [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4557

Pages