Socially Pertinent Robots in Gerontological Healthcare

Livrables

User feedback from the intermediate validation (realistic/relevant environments)

Result of T14 Report the user feedback obtained on the intermediate experiments on both realistic and relevant environments

Initial Advisory Board recommendations

Document reporting the recommendations of the Advisory Board after its first meeting

Intermediate Advisory Board recommendations

Document reporting the recommendations of the Advisory Board after its second meeting

Specifications of the generator of situated interactions

Initial report of T62 Generation of Multiparty Situated Interactions In this report we will provide the specifications for the generator of interactive data

Mid-point report on organised scientific events and future plans

We will summarize the dissemination and communication actions taken in scientific events and describe the plan for the rest of the project duration

User feedback from the preliminary validation (realistic environments)

Result of T13 Report the user feedback from the various experiments ran on realistic environments laboratory

Dissemination and communication strategy

Report the communication plan and plan the various actions to be taken to promote the project as well as its results in the targeted sectors

Privacy and Ethics guidelines for experimental validation and data collection

The deliverable will contain SPRING necessary documents for handling Ethics and Privacy issues For instance approval of the local Ethical committees and GDRP compliance report of all experimental validation and data collection in the project

Human description in relevant environments

Result of T41 Fully functional framework for human face and body analysis from visual data It includes tools for multitarget body pose estimation and face analysis tested in relevant environments The framework incorporate information from audiobased speaker recognition

Robot non-verbal behaviour system in target environments

Will deliver the final software design and implementation of the robot non-verbal behaviour manager for the target environment. This includes the interface between the non-verbal behaviour manager, the task planner, and the conversational system.

Multi-party conversational system in target environments

Will deliver the final software components for multi-party interaction including the conversational system, NLU, dialogue management, and NLG geared towards dealing with realistic scenarios (T5.3).

Audio-visual speaker separation/tracking with a static robot

Results of T3.2. and T3.3 A software package incorporating video localization information to assist the audio diarisation and separation tasks for the use of the robot’s dialogue system in both static and dynamic scenarios.

Audio speaker diarisation and extraction with a moving robot

Result of T3.3. A software package, implemented for moving robot that accounts for the dynamics of the acoustic scenario, to disarise and extract the desired speaker.

Semantics-based localisation in realistic environments

Research results and implementation of building a geometric and semantic maps for visual localization developed in T21

Neural network architecture specification and design

This deliverable will provide the software basis for the development of the neural architectures used in the project. After an initial phase of collecting requirements for these networks, we will provide some basis architectures from which partners can build their systems.

Software for generating multi-party situated interactions

Result of T62 This deliverable will present the software for generating multiparty situated interactions

Human description in realistic environments

Result of T41 Research results and implementation of multitarget body pose and state estimation based on deep and transfer learning and domain adaptation techniques

Robot non-verbal behaviour system in realistic environments

Will deliver the initial software design and implementation of the robot nonverbal behaviour manager for a realistic environment This includes the interface between the nonverbal behaviour manager the task planner and the conversational system

Initial software architecture for SPRING-REEM

This Deliverable contains the initial SW architecture (modules and applications integration), after the preliminary Software Integration Cycle.

Visual-based localisation in realistic environments

Result of T21 Research results of visual localization of a robot in realistic indoor environments based on viewpoint visual similarity and via geometric matching and careful camera pose verification using camera pose computation in 3D or similar weaker geometrical verification techniques interaction with the lowlevel robot behaviour and robots odometry

Semantics-based localisation in relevant environments

Will deliver software with a Natural Language vocabulary (categories) for object recognition that interfaces with the high-level task scheduler for semantic scene understanding (T2.3) and builds towards interfacing the Interaction Manager to incorporate object affordances (T2.4).

Mature software architecture for SPRING-REEM

This Deliverable contains the mature SW architecture (modules and applications integration), after the Intermediate Software Integration Cycle.

Visual-based localisation in relevant environments

Result of T24 Implementation of visual localization of a robot in realistic indoor environments based on viewpoint visual similarity and via geometric matching and careful camera pose verification using camera pose computation in 3D or similar weaker geometrical verification techniques interaction with the lowlevel robot behaviour and robots odometry

Learning scene representations in realistic environments

Result of T24 It includes a first prototype implementation of tools for modeling scenes eg objects affordances people behaviors in order to support the semantic localization module from 21

Audio-visual speaker tracking in realistic environments

Midway result of T31 AudioVisual Speaker Detection and Tracking This deliverable will present the software to detect and track speakers in realistic environments

Audio-visual speaker tracking in relevant environments

Endway result of T31 AudioVisual Speaker Detection and Tracking This deliverable will present the software to detect and track speakers in relevant environments

High-Level task planner in relevant environments

Software package interleaving the robot application (direct interaction with users) and the low level modules in relevant environments.

Audio-visual data simulator

Result of task T22 AudioVisual data simulator A synthetic data simulator will be developed in order able to generate realistic auditory and visual data This simulator will build upon the interactive data collected in T12

Multi-modal behaviour recognition in realistic environments

Result of T42 First prototype implementation of tools for single target behaviour recognition and for grouplevel behaviour analysis integrating information about face and body motions derived from audiovisual data T41 objects in the scene T23 and proxemics features

Multi-modal behaviour recognition in relevant environments

Result of T4.2. Fully functional human behaviour recognition framework including advanced components for individual- and group-level analysis validated in relevant environment.

Learning scene representations in relevant environments

"""Result of T2.3. Implementation of learning a scene representation for facilitation localization of a robot in realistic indoor environments."

Initial high-level task planner and conversational system prototype for realistic environments.

Will deliver software serving as the initial conversational system and highlevel task planner prototype for realistic environments T51

Audio speaker diarisation and extraction with a static robot

Result of T3.2. A software package, implemented for static robot, for audio speaker diarisation and extraction.

Multi-party ASR and conversational system in realistic environments

Result of T53 A software package for multiparty ASR and conversational system in realistic environments using preliminary speaker separation algorithm

Final neural network architectures

Result of T6.1. This deliverable will present the software of the final neural architectures used in SPRING.

Website and online presence

Result of T81 We will present the various actions taken aiming to ensure the online presence of SPRING basis for the dissemination and communication of the associated results

Publications

Intérêt de la robotique sociale et d’assistance auprès des sujets âgés (Interest in social and assistive robotics for elderly subjects)

Auteurs: Maribel Pino, Sébastien Dacunha , Étienne Berger, Anna Goncalves , Anne-Sophie Rigaud
Publié dans: Actualités Pharmaceutiques, Numéro 60, 611, 2021, Page(s) 36-39, ISSN 0515-3700
Éditeur: Elsevier BV
DOI: 10.1016/j.actpha.2021.10.010

Globally Optimal Solution to Inverse Kinematics of 7DOF Serial Manipulator

Auteurs: Pavel Trutman; Mohab Safey El Din; Didier Henrion; Tomas Pajdla
Publié dans: IEEE Robotics and Automation Letters, 2022, ISSN 2377-3766
Éditeur: IEEE
DOI: 10.48550/arxiv.2007.12550

SocialInteractionGAN: Multi-person Interaction Sequence Generation

Auteurs: Airale, Louis; Vaufreydaz, Dominique; Alameda-Pineda, Xavier
Publié dans: Transactions on Automatic Control, 2021, ISSN 1558-2523
Éditeur: IEEE

Audio source separation by activity probability detection with maximum correlation and simplex geometry

Auteurs: Bracha Laufer-Goldshtein; Ronen Talmon; Sharon Gannot
Publié dans: EURASIP Journal on Audio, Speech, and Music Processing, Numéro 2021, 5, 2021, ISSN 1687-4722
Éditeur: Springer
DOI: 10.1186/s13636-021-00195-7

Variational Meta Reinforcement Learning for Social Robotics

Auteurs: Ballou, Anand; Alameda-Pineda, Xavier; Reinke, Chris
Publié dans: Applied Intelligence, 2023, ISSN 0924-669X
Éditeur: Kluwer Academic Publishers
DOI: 10.48550/arxiv.2206.03211

Galois/monodromy groups for decomposing minimal problems in 3D reconstruction

Auteurs: Duff, Timothy; Korotynskiy, Viktor; Pajdla, Tomas; Regan, Margaret H.
Publié dans: SIAM Journal on Applied Algebra and Geometry, 2022, ISSN 2470-6566
Éditeur: Society for Industrial and Applied Mathematics
DOI: 10.1137/21m142287

The hybrid Cramér-Rao lower bound for simultaneous self-localization and room geometry estimation

Auteurs: Maya Veisman; Yair Noam; Sharon Gannot
Publié dans: EURASIP Journal on Advances in Signal Processing, Numéro Vol 2021, Iss 1, Pp 1-22 (2021), 2021, ISSN 1687-6180
Éditeur: Springer
DOI: 10.1186/s13634-020-00702-6

Variational Inference and Learning of Piecewise linear Dynamical Systems.

Auteurs: Xavier Alameda-Pineda; Vincent Drouard; Radu Horaud
Publié dans: IEEE Transactions on Neural Networks and Learning Systems, Numéro 2020, 2020, ISSN 2162-2388
Éditeur: IEEE
DOI: 10.1109/tnnls.2021.3054407

Orthogonal SVD Covariance Conditioning and Latent Disentanglement

Auteurs: Yue Song, Nicu Sebe, Wei Wang
Publié dans: IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, ISSN 0162-8828
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tpami.2022.3228979

Unsupervised High-Resolution Portrait Gaze Correction and Animation

Auteurs: Jichao Zhang, Jingjing Chen, Hao Tang, Enver Sangineto, Peng Wu, Yan Yan, Nicu Sebe, Wei Wang
Publié dans: IEEE Transactions on Image Processing, 2022, ISSN 1057-7149
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tip.2022.3191852

Learning and controlling the source-filter representation of speech with a variational autoencoder

Auteurs: Samir Sadok; Simon Leglaive; Laurent Girin; Xavier Alameda-Pineda; Renaud Séguier
Publié dans: Speech Communication, Numéro 2, 2023, ISSN 0167-6393
Éditeur: Elsevier BV
DOI: 10.48550/arxiv.2204.07075

Semi-Supervised Source Localization in Reverberant Environments With Deep Generative Modeling

Auteurs: Michael J. Bianco; Sharon Gannot; Efren Fernandez-Grande; Peter Gerstoft
Publié dans: IEEE Access, Numéro Vol 9, Pp 84956-84970 (2021), 2021, ISSN 2169-3536
Éditeur: Institute of Electrical and Electronics Engineers Inc.
DOI: 10.1109/access.2021.3087697

Multi-frame Motion Segmentation by Combining Two-Frame Results

Auteurs: Federica Arrigoni; Elisa Ricci; Tomas Pajdla
Publié dans: IEEE Transactions on Image Processing, 2022, ISSN 0920-5691
Éditeur: Kluwer Academic Publishers
DOI: 10.1007/s11263-021-01544-x

Multi-frame Motion Segmentation by Combining Two-Frame Results

Auteurs: Federica Arrigoni, Elisa Ricci & Tomas Pajdla
Publié dans: International Journal of Computer Vision, 2023, ISSN 0920-5691
Éditeur: Kluwer Academic Publishers

Forward-backward recursive expectation-maximization for concurrent speaker tracking

Auteurs: Dorfan, Y., Schwartz, B. & Gannot, S
Publié dans: EURASIP Journal on Audio, Speech, and Music Processing, Numéro 2021, 2 (2021), 2021, ISSN 1687-4722
Éditeur: Springer
DOI: 10.1186/s13636-020-00189-x

TransCenter: Transformers With Dense Representations for Multiple-Object Tracking

Auteurs: Yihong Xu; Yutong Ban; Guillaume Delorme; Chuang Gan; Daniela Rus; Xavier Alameda-Pineda
Publié dans: IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, ISSN 0162-8828
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tpami.2022.3225078

SocialInteractionGAN: Multi-person Interaction Sequence Generation

Auteurs: Louis Airale; Dominique Vaufreydaz; Xavier Alameda-Pineda
Publié dans: IEEE Transactions on Affective Computing, 2021, ISSN 1949-3045
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.48550/arxiv.2103.05916

Successor Feature Representations

Auteurs: Reinke, Chris; Alameda-Pineda, Xavier
Publié dans: Transactions on Machine Learning Research, 2023, ISSN 2835-8856
Éditeur: OpenReview
DOI: 10.48550/arxiv.2110.15701

Dynamical Variational Autoencoders: A Comprehensive Review

Auteurs: Laurent Girin, Simon Leglaive , Xiaoyu Bie , Julien Diard , Thomas Hueber , Xavier Alameda-Pineda
Publié dans: Foundations and Trends in Machine Learning, Numéro 15, 1-2, 2021, ISSN 1935-8237
Éditeur: Now Publishers Inc.
DOI: 10.1561/2200000089

Dynamically localizing multiple speakers based on the time-frequency domain

Auteurs: Hodaya Hammer; Shlomo E. Chazan; Jacob Goldberger; Sharon Gannot
Publié dans: EURASIP Journal on Audio, Speech, and Music Processing, Numéro Vol 2021, Iss 1, Pp 1-10 (2021), 2021, ISSN 1687-4722
Éditeur: Springer
DOI: 10.1186/s13636-021-00203-w

Uncertainty-aware Contrastive Distillation for Incremental Semantic Segmentation

Auteurs: Guanglei Yang; Enrico Fini; Dan Xu; Paolo Rota; Mingli Ding; Moin Nabi; Xavier Alameda-Pineda; Elisa Ricci
Publié dans: IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, ISSN 0162-8828
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.48550/arxiv.2203.14098

Semi-Supervised Multiple Source Localization Using Relative Harmonic Coefficients Under Noisy and Reverberant Environments

Auteurs: Y. Hu, P. N. Samarasinghe, S. Gannot and T. D. Abhayapala
Publié dans: IEEE/ACM Transactions on Audio, Speech, and Language Processing, Numéro vol. 28, 2020, Page(s) pp. 3108-3123, ISSN 2329-9304
Éditeur: IEEE
DOI: 10.1109/taslp.2020.3037521

Variational Structured Attention Networks for Deep Visual Representation Learning

Auteurs: Guanglei Yang; Paolo Rota; Xavier Alameda-Pineda; Dan Xu; Mingli Ding; Elisa Ricci
Publié dans: IEEE Transactions on Image Processing, 2022, ISSN 1057-7149
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tip.2021.3137647

Continual Attentive Fusion for Incremental Learning in Semantic Segmentation

Auteurs: Guanglei Yang; Enrico Fini; Dan Xu; Paolo Rota; Mingli Ding; Tang Hao; Xavier Alameda-Pineda; Elisa Ricci
Publié dans: IEEE Transactions on Multimedia, 2022, ISSN 1520-9210
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.48550/arxiv.2202.00432

Training-Based Multiple Source Tracking Using Manifold-Learning and Recursive Expectation-Maximization

Auteurs: Avital Bross; Sharon Gannot
Publié dans: IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023, ISSN 2329-9290
Éditeur: IEEE Advancing Technology for Humanity
DOI: 10.1109/taslp.2023.3245414

Expression-preserving face frontalization improves visually assisted speech processing

Auteurs: Zhiqi Kang; Mostafa Sadeghi; Radu Horaud; Xavier Alameda-Pineda
Publié dans: International Journal of Computer Vision, 2023, ISSN 0920-5691
Éditeur: Kluwer Academic Publishers
DOI: 10.48550/arxiv.2204.02810

Simultaneous Tracking and Separation of Multiple Sources Using Factor Graph Model

Auteurs: Koby Weisberg; Bracha Laufer-Goldshtein; Sharon Gannot
Publié dans: IEEE/ACM Transactions on Audio, Speech, and Language Processing, Numéro 28, 2020, Page(s) 2848-2864, ISSN 2329-9290
Éditeur: IEEE Advancing Technology for Humanity
DOI: 10.1109/taslp.2020.3028650

Estimation of acoustic echoes using expectation-maximization methods

Auteurs: Saqib, U., Gannot, S. & Jensen, J.
Publié dans: EURASIP Journal on Audio, Speech, and Music Processing, Numéro 2020, 12 (2020), 2020, ISSN 1687-4722
Éditeur: Springer
DOI: 10.1186/s13636-020-00179-z

Unsupervised Speech Enhancement using Dynamical Variational Autoencoders

Auteurs: Xiaoyu Bie; Simon Leglaive; Xavier Alameda-Pineda; Laurent Girin
Publié dans: IEEE/ACM Transactions on Audio, Speech and Language Processing, 2022, ISSN 2329-9290
Éditeur: IEEE Advancing Technology for Humanity
DOI: 10.1109/taslp.2022.3207349

A recursive expectation-maximization algorithm for speaker tracking and separation

Auteurs: Schwartz, O., Gannot, S.
Publié dans: EURASIP Journal on Audio, Speech, and Music Processing, Numéro 2021, 43, 2021, ISSN 1687-4722
Éditeur: Springer
DOI: 10.1186/s13636-021-00228-1

An online algorithm for echo cancellation, dereverberation and noise reduction based on a Kalman-EM Method

Auteurs: Nili Cohen; Gershon Hazan; Boaz Schwartz; Sharon Gannot
Publié dans: EURASIP Journal on Audio, Speech, and Music Processing, Numéro 2021, 33, 2021, ISSN 1687-4722
Éditeur: Springer
DOI: 10.1186/s13636-021-00219-2

Mixture of Inference Networks for VAE-based Audio-visual Speech Enhancement

Auteurs: M. Sadeghi and X. Alameda-Pineda
Publié dans: IEEE Transactions on Signal Processing, Numéro vol. 69, 2021, Page(s) pp. 1899-1909, ISSN 1941-0476
Éditeur: IEEE
DOI: 10.1109/tsp.2021.3066038

Viewing Graph Solvability via Cycle Consistency

Auteurs: Federica Arrigoni, Andrea Fusiello, Elisa Ricci, Tomas Pajdla
Publié dans: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Numéro 2021, 2021, Page(s) 5540-5549
Éditeur: IEEE

A Multimodal Dynamical Variational Autoencoder for Audiovisual Speech Representation Learning

Auteurs: Sadok, Samir; Leglaive, Simon; Girin, Laurent; Alameda-Pineda, Xavier; Séguier, Renaud
Publié dans: M4MM '22: Proceedings of the 1st International Workshop on Methodologies for Multimedia, 2022
Éditeur: ACM
DOI: 10.48550/arxiv.2305.03582

Array Configuration Mismatch in Deep DOA Estimation: Towards Robust Training

Auteurs: Ayal Schwartz; Elior Hadad; Sharon Gannot; Shlomo E. Chazan
Publié dans: 2023 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), 2023
Éditeur: IEEE
DOI: 10.1109/waspaa58266.2023.10248048

Vocabulary-free Image Classification

Auteurs: Conti, Alessandro; Fini, Enrico; Mancini, Massimiliano; Rota, Paolo; Wang, Yiming; Ricci, Elisa
Publié dans: NeurIPS2023, Numéro 1, 2024
Éditeur: NeurIPS
DOI: 10.48550/arxiv.2306.00917

Uncertainty-Guided Source-Free Domain Adaptation

Auteurs: Subhankar Roy, Martin Trapp, Andrea Pilzer, Juho Kannala, Nicu Sebe, Elisa Ricci, Arno Solin
Publié dans: ECCV 2022, 2022
Éditeur: ECVA
DOI: 10.48550/arxiv.2208.07591

Multi-Person Extreme Motion Prediction

Auteurs: Wen Guo; Xiaoyu Bie; Xavier Alameda-Pineda; Francesc Moreno-Noguer
Publié dans: IEEE/CVF Conference on Computer Vision and Pattern Recognition, Numéro 2, 2022
Éditeur: IEEE
DOI: 10.48550/arxiv.2105.08825

Intrinsic-Extrinsic Preserved GANs for Unsupervised 3D Pose Transfer

Auteurs: Haoyu Chen, Hao Tang, Henglin Shi, Wei Peng, Nicu Sebe, Guoying Zhao
Publié dans: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Numéro 2021, 2021, Page(s) 8630-8639
Éditeur: IEEE

Developing a Social Conversational Robot for the Hospital waiting room

Auteurs: Nancie Gunson, Daniel Hernandez Garcia, Weronika Sieinska, Christian Dondrup, Oliver Lemon
Publié dans: RO-MAN 2022, 2022
Éditeur: IEEE
DOI: 10.1109/ro-man53752.2022.9900827

Novel Class Discovery in Semantic Segmentation

Auteurs: Yuyang Zhao, Zhun Zhong, Nicu Sebe, Gim Hee Lee
Publié dans: 2022
Éditeur: CVF
DOI: 10.48550/arxiv.2112.01900

Neighborhood Contrastive Learning for Novel Class Discovery

Auteurs: Zhong, Zhun; Fini, Enrico; Subhankar Roy; Zhiming Luo; Ricci, Elisa; Sebe, Nicu
Publié dans: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 2021, 2021, Page(s) 10867-10875
Éditeur: IEEE
DOI: 10.5281/zenodo.5014108

Transformer-Based Attention Networks for Continuous Pixel-Wise Prediction

Auteurs: Guanglei Yang, Hao Tang, Mingli Ding, Nicu Sebe, Elisa Ricci
Publié dans: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Numéro 2021, 2021, Page(s) 16269-16279
Éditeur: IEEE

A Unified Objective for Novel Class Discovery

Auteurs: Enrico Fini, Enver Sangineto, Stéphane Lathuilière, Zhun Zhong, Moin Nabi, Elisa Ricci
Publié dans: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Numéro 2021, 2021, Page(s) 9284-9292
Éditeur: IEEE

Combining Visual and Social Dialogue for Human-Robot Interaction

Auteurs: Nancie Gunson; Daniel Hernandez Garcia; Jose L. Part; Yanchao Yu; Weronika Sieińska; Christian Dondrup; Oliver Lemon
Publié dans: ICMI '21: Proceedings of the 2021 International Conference on Multimodal Interaction, Numéro 2021, 2021, Page(s) 841-842
Éditeur: ICMI
DOI: 10.1145/3462244.3481303

Cluster-level pseudo-labelling for source-free cross-domain facial expression recognition

Auteurs: Alessandro Conti; Paolo Rota; Yiming Wang; Elisa Ricci
Publié dans: BMCV 2022, 2022
Éditeur: The British Machine Vision Association and Society for Pattern Recognition
DOI: 10.5281/zenodo.7296310

RankFeat: Rank-1 Feature Removal for Out-of-distribution Detection

Auteurs: Yue Song, Nicu Sebe, Wei Wang
Publié dans: NeurIPS 2022, 2022
Éditeur: OpenReview
DOI: 10.48550/arxiv.2209.08590

Multiple Speaker Localization using Mixture of Gaussian Model with Manifold-based Centroids

Auteurs: A. Bross, B. Laufer-Goldshtein and S. Gannot
Publié dans: 2020 28th European Signal Processing Conference (EUSIPCO), Numéro 2020, 2021, Page(s) 895-899
Éditeur: IEEE
DOI: 10.23919/eusipco47968.2020.9287796

OpenMix: Reviving Known Knowledge for Discovering Novel Visual Categories in an Open World

Auteurs: Zhun Zhong, Linchao Zhu, Zhiming Luo, Shaozi Li, Yi Yang, Nicu Sebe
Publié dans: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021
Éditeur: IEEE

Misalignment Recognition in Acoustic Sensor Networks Using a Semi-Supervised Source Estimation Method and Markov Random Fields

Auteurs: Miller, Gabriel F; Brendel, Andreas; Kellermann, Walter; Gannot, Sharon
Publié dans: ICASSP, Numéro 2021, 2021
Éditeur: ICASSP
DOI: 10.1109/icassp39728.2021.9413765

Self-Supervised Models are Continual Learners

Auteurs: Fini, Enrico; da Costa, Victor G. Turrisi; Alameda-Pineda, Xavier; Ricci, Elisa; Alahari, Karteek; Mairal, Julien
Publié dans: CVPR 2022 - IEEE/CVF Conference on Computer Vision and Pattern Recognition, Numéro 6, 2022
Éditeur: CVF
DOI: 10.48550/arxiv.2112.04215

A Bayesian Hierarchical Mixture of Gaussian Model for Multi-Speaker DOA Estimation and Separation

Auteurs: Y. Laufer and S. Gannot
Publié dans: 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing (MLSP), Numéro 2020, 2020, Page(s) 1-6
Éditeur: IEEE
DOI: 10.1109/mlsp49062.2020.9231852

Curriculum Graph Co-Teaching for Multi-Target Domain Adaptation

Auteurs: Subhankar Roy, Evgeny Krivosheev, Zhun Zhong, Nicu Sebe, Elisa Ricci
Publié dans: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 2021, 2021, Page(s) 5351-5360
Éditeur: IEEE

Probabilistic Fusion of Persons' Body Features: The Mr. Potato Algorithm

Auteurs: Séverin Lemaignan, Lorenzo Ferrini
Publié dans: HRI '24: Proceedings of the 2024 ACM/IEEE International Conference on Human-Robot Interaction, 2024
Éditeur: ACM
DOI: 10.1145/3610977.3637479

OpenMix: Reviving Known Knowledge for Discovering Novel Visual Categories in an Open World

Auteurs: Zhun Zhong; Linchao Zhu; Zhiming Luo; Shaozi Li; Yi Yang; Nicu Sebe
Publié dans: CVPR 2021, 2021
Éditeur: CVF
DOI: 10.1109/cvpr46437.2021.00934

ROS4HRI: Standardising an Interface for Human-Robot Interaction

Auteurs: Raquel Ros, Séverin Lemaignan, Lorenzo Ferrini, Antonio Andriella, Aina Irisarri
Publié dans: HRI 2023 Workshop, 2023
Éditeur: IEEE

From two rolling shutters to one global shutter

Auteurs: Albl, Cenek; Kukelova, Zuzana; Larsson, Viktor; Pajdla, Tomas; Schindler, Konrad
Publié dans: 2020
Éditeur: CVF
DOI: 10.48550/arxiv.2006.01964

Viewing Graph Solvability via Cycle Consistency

Auteurs: Federica Arrigoni, Andrea Fusiello, Elisa Ricci, Tomas Pajdla
Publié dans: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2022
Éditeur: IEEE
DOI: 10.1109/iccv48922.2021.00549

Open-source Natural Language Processing on the PAL Robotics ARI Social Robot

Auteurs: Séverin Lemaignan, Sara Cooper, Raquel Ros, Lorenzo Ferrini, Antonio Andriella, Aina Irisarri
Publié dans: HRI '23: Companion of the 2023 ACM/IEEE International Conference on Human-Robot Interaction, 2023
Éditeur: ACM

3D-Aware Semantic-Guided Generative Model for Human Synthesis

Auteurs: Jichao Zhang, Enver Sangineto, Hao Tang, Aliaksandr Siarohin, Zhun Zhong, Nicu Sebe, Wei Wang
Publié dans: ECCV 2022, 2022
Éditeur: ECVA
DOI: 10.48550/arxiv.2112.01422

Unifying Bottom-Up and Top-Down Attention Models for Social Robots

Auteurs: S. Lemaignan; S. Cooper; R. Ros; L. Ferrini; A. Andriella; A. Irisarri
Publié dans: HRI 2023, 2023
Éditeur: ACM
DOI: 10.1145/3568294.3580041

Semi-Supervised Source Localization with Deep Generative Modeling

Auteurs: Michael J. Bianco; Sharon Gannot; Peter Gerstoft
Publié dans: 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing (MLSP), 2021
Éditeur: IEEE
DOI: 10.1109/mlsp49062.2020.9231825

Dataset and Evaluation of Automatic Speech Recognition for Multi-lingual Intent Recognition on Social Robots

Auteurs: Antonio Andriella, Raquel Ros, Yoav Ellinson, Sharon Gannot, Séverin Lemaignan
Publié dans: HRI '24: Proceedings of the 2024 ACM/IEEE International Conference on Human-Robot Interaction, 2024
Éditeur: ACM
DOI: 10.1145/3610977.3637473

Intrinsic-Extrinsic Preserved GANs for Unsupervised 3D Pose Transfer

Auteurs: Haoyu Chen; Hao Tang; Henglin Shi; Wei Peng; Nicu Sebe; Guoying Zhao
Publié dans: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021
Éditeur: CVF
DOI: 10.48550/arxiv.2108.07520

Learning to Generalize Unseen Domains via Memory-based Multi-Source Meta-Learning for Person Re-Identification

Auteurs: Yuyang Zhao, Zhun Zhong, Fengxiang Yang, Zhiming Luo, Yaojin Lin, Shaozi Li, Nicu Sebe
Publié dans: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 2021, 2021, Page(s) 6277-6286
Éditeur: IEEE

Semi-supervised learning made simple with self-supervised clustering

Auteurs: Fini, Enrico; Astolfi, Pietro; Alahari, Karteek; Alameda-Pineda, Xavier; Mairal, Julien; Nabi, Moin; Ricci, Elisa
Publié dans: 2023
Éditeur: CVF
DOI: 10.48550/arxiv.2306.07483

Minimal Rolling Shutter Absolute Pose with Unknown Focal Length and Radial Distortion

Auteurs: Kukelova, Zuzana; Albl, Cenek; Sugimoto, Akihiro; Schindler, Konrad; Pajdla, Tomas
Publié dans: European Conference on Computer Vision, Numéro 2020, 2020, Page(s) 698-714
Éditeur: Springer
DOI: 10.5281/zenodo.4335228

Why Approximate Matrix Square Root Outperforms Accurate SVD in Global Covariance Pooling?

Auteurs: Yue Song, Nicu Sebe, Wei Wang
Publié dans: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Numéro 2021, 2021, Page(s) 1115-1123
Éditeur: IEEE

Deep Ranking-Based DOA Tracking Algorithm

Auteurs: R. Opochinsky, G. Chechik and S. Gannot
Publié dans: 2021 29th European Signal Processing Conference (EUSIPCO), Numéro 2021, 2021, Page(s) 1020-1024
Éditeur: IEEE
DOI: 10.23919/eusipco54536.2021.9616297

Optimizing Elimination Templates by Greedy Parameter Search

Auteurs: Evgeniy Martyushev, Jana Vrablikova, Tomas Pajdla
Publié dans: 2022
Éditeur: CVF
DOI: 10.48550/arxiv.2203.14901

Uncertainty Based Camera Model Selection

Auteurs: Michal Polic, Stanislav Steidl, Cenek Albl, Zuzana Kukelova, Tomas Pajdla
Publié dans: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 2020, 2020, Page(s) 5991-6000
Éditeur: IEEE

A Bayesian Hierarchical Model for Blind Audio Source Separation

Auteurs: Y. Laufer and S. Gannot
Publié dans: 2020 28th European Signal Processing Conference (EUSIPCO), Numéro 2020, 2021, Page(s) 276-280
Éditeur: IEEE
DOI: 10.23919/eusipco47968.2020.9287348

Speech enhancement with mixture-of-deep-experts with clean clustering pre-training

Auteurs: Chazan, Shlomo E.; Goldberger, Jacob; Gannot, Sharon
Publié dans: ICASSP, Numéro 2021, 2021
Éditeur: ICASSP

Single microphone speaker extraction using unified time-frequency Siamese-Unet

Auteurs: Aviad Eisenberg; Sharon Gannot; Shlomo E. Chazan
Publié dans: 2022 30th European Signal Processing Conference (EUSIPCO), 2022
Éditeur: IEEE
DOI: 10.23919/eusipco55093.2022.9909545

Novel Class Discovery in Semantic Segmentation

Auteurs: Yuyang Zhao; Zhun Zhong; Nicu Sebe; Gim Hee Lee
Publié dans: CVPR 2022, 2022
Éditeur: Computer Vision Foundation
DOI: 10.5281/zenodo.7100314

Multimodal Emotion Recognition with Modality-Pairwise Unsupervised Contrastive Loss

Auteurs: Riccardo Franceschini, Enrico Fini, Cigdem Beyan, Alessandro Conti, Federica Arrigoni, Elisa Ricci
Publié dans: ICPR 2022, 2022
Éditeur: IEEE
DOI: 10.1109/icpr56361.2022.9956589

The impact of removing head movements on audio-visual speech enhancement

Auteurs: Zhiqi Kang, Mostafa Sadeghi, Radu Horaud, Xavier Alameda-Pineda, Jacob Donley, Anurag Kumar
Publié dans: ICASSP 2022, 2022
Éditeur: IEEE
DOI: 10.1109/icassp43922.2022.9746401

ARI: the Social Assistive Robot and Companion

Auteurs: Sara Cooper; Alessandro Di Fava; Carlos Vivas; Luca Marchionni; Francesco Ferro
Publié dans: 2020 29th IEEE International Conference on Robot and Human Interactive Communication (RO-MAN), 2020
Éditeur: IEEE
DOI: 10.1109/ro-man47096.2020.9223470

Blind Audio Source Separation Using Two Expectation-Maximization Algorithms

Auteurs: A. Eisenberg, B. Schwartz and S. Gannot
Publié dans: 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing (MLSP), Numéro 2020, 2020, Page(s) 1-6
Éditeur: IEEE
DOI: 10.1109/mlsp49062.2020.9231931

Evaluation and Design Guidelines for Combining Open-Domain Social Conversation with Task-Based Dialogue in Intelligent Buildings

Auteurs: Nancie Gunson; Weronika Sieinska; Christopher Walsh; Christian Dondrup; Oliver Lemon
Publié dans: IVA '20: Proceedings of the 20th ACM International Conference on Intelligent Virtual Agents, Numéro 2020, 2020
Éditeur: Association for Computing Machinery
DOI: 10.1145/3383652.3423889

Online Blind Audio Source Separation using Recursive Expectation-Maximization

Auteurs: Aviad Eisenberg, Boaz Schwartz and Sharon Gannot
Publié dans: INTERSPEECH 2021, 2021
Éditeur: ISCA
DOI: 10.21437/interspeech.2021-662

A Visually-Aware Conversational Robot Receptionist

Auteurs: Nancie Gunson, Daniel Hernandez Garcia, Weronika Sieińska, Angus Addlesee, Christian Dondrup, Oliver Lemon, Jose L. Part, Yanchao Yu
Publié dans: SIGdial 2022, 2022
Éditeur: ACL

Scene-Agnostic Multi-Microphone Speech Dereverberation

Auteurs: Yochai Yemini, Ethan Fetaya, Haggai Maron, Sharon Gannot
Publié dans: INTERSPEECH 2021, Numéro 2021, 2021
Éditeur: ISCA
DOI: 10.21437/interspeech.2021-889

Class-incremental Novel Class Discovery

Auteurs: Subhankar Roy, Mingxuan Liu, Zhun Zhong, Nicu Sebe, Elisa Ricci
Publié dans: ECCV 2022, 2022
Éditeur: ECVA
DOI: 10.48550/arxiv.2207.08605

Explainable Representations of the Social State: A Model for Social Human-Robot Interactions

Auteurs: García, Daniel Hernández; Yu, Yanchao; Sieińska, Weronika; Part, Jose L.; Gunson, Nancie; Lemon, Oliver; Dondrup, Christian
Publié dans: AAAI FSS-20 AI-HRI, 2020
Éditeur: Association for the Advancement of Artificial Intelligence

Multimodal Across Domains Gaze Target Detection

Auteurs: Francesco Tonini; Cigdem Beyan; Elisa Ricci
Publié dans: ICMI '22: Proceedings of the 2022 International Conference on Multimodal Interaction, 2022
Éditeur: ACM
DOI: 10.48550/arxiv.2208.10822

Unsupervised Domain Adaptation for Video Transformers in Action Recognition

Auteurs: Victor G. Turrisi da Costa, Giacomo Zara, Paolo Rota, Thiago Oliveira-Santos, Nicu Sebe, Vittorio Murino, Elisa Ricci
Publié dans: ICPR 2022, 2022
Éditeur: IEEE
DOI: 10.48550/arxiv.2207.12842

On The Importance of Acoustic Reflections in Beamforming

Auteurs: Oren Shmaryahu; Sharon Gannot
Publié dans: IWAENC 2022, 2022
Éditeur: IEE
DOI: 10.1109/iwaenc53105.2022.9914749

Learning to Generalize Unseen Domains via Memory-based Multi-Source Meta-Learning for Person Re-Identification

Auteurs: Yuyang Zhao; Zhun Zhong; Fengxiang Yang; Zhiming Luo; Yaojin Lin; Shaozi Li; Nicu Sebe
Publié dans: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, 2021
Éditeur: CVF
DOI: 10.5281/zenodo.5014450

It's Good to Chat?

Auteurs: Gunson, Nancie; Sieińska, Weronika; Walsh, Christopher; Dondrup, Christian; Lemon, Oliver
Publié dans: isbn: 9781450375863, Numéro 1, 2020
Éditeur: Association for Computing Machinery

Synchronization of Group-Labelled Multi-Graphs

Auteurs: Andrea Porfiri Dal Cin, Luca Magri, Federica Arrigoni, Andrea Fusiello, Giacomo Boracchi
Publié dans: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Numéro 2021, 2021, Page(s) 6453-6463
Éditeur: IEEE

The Impact of Removing Head Movements on Audio-visual Speech Enhancement

Auteurs: Zhiqi Kang; Mostafa Sadeghi; Radu Horaud; Xavier Alameda-Pineda; Jacob Donley; Anurag Kumar
Publié dans: 2022
Éditeur: ICASSP
DOI: 10.48550/arxiv.2202.00538

Making Affine Correspondences Work in Camera Geometry Computation

Auteurs: Barath, Daniel; Polic, Michal; Förstner, Wolfgang; Sattler, Torsten; Pajdla, Tomas; Kukelova, Zuzana
Publié dans: European Conference on Computer Vision, Numéro 2020, 2020, Page(s) 723-740
Éditeur: Springer
DOI: 10.1007/978-3-030-58621-8_42

Fast Differentiable Matrix Square Root

Auteurs: Yue Song, Nicu Sebe, Wei Wang
Publié dans: ICLR 2022, 2023
Éditeur: OpenReview
DOI: 10.48550/arxiv.2201.08663

MultiBodySync: Multi-Body Segmentation and Motion Estimation via 3D Scan Synchronization

Auteurs: Jiahui Huang, He Wang, Tolga Birdal, Minhyuk Sung, Federica Arrigoni, Shi-Min Hu, Leonidas J. Guibas
Publié dans: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 2021, 2021, Page(s) 7108-7118
Éditeur: IEEE

A Benchmark of Dynamical Variational Autoencoders applied to Speech Spectrogram Modeling

Auteurs: Bie, Xiaoyu; Girin, Laurent; Leglaive, Simon; Hueber, Thomas; Alameda-Pineda, Xavier
Publié dans: Interspeech 2021 - 22nd Annual Conference of the International Speech Communication Association, Aug 2021, Brno, Czech Republic. pp.1-5, Numéro 2021, 2021, Page(s) 46-50
Éditeur: International Speech Communication Association (ISCA)
DOI: 10.21437/interspeech.2021-256

Motion Segmentation with Pairwise Matches and Unknown Number of Motions

Auteurs: Federica Arrigoni, Luca Magri, and Tomas Pajdla
Publié dans: 2020 25th International Conference on Pattern Recognition (ICPR), Numéro 2020, 2020
Éditeur: IEEE
DOI: 10.1109/icpr48806.2021.9413142

On the Usage of the Trifocal Tensor in Motion Segmentation

Auteurs: Federica Arrigoni, Luca Magri, Tomas Pajdla
Publié dans: ECCV’20 Online, 2020
Éditeur: European Computer Vision Association (ECVA)

Learning Visual Voice Activity Detection with an Automatically Annotated Dataset

Auteurs: Guy, Sylvain; Lathuilière, Stéphane; Mesejo, Pablo; Horaud, Radu
Publié dans: ICPR 2020 - 25th International Conference on Pattern Recognition, Jan 2021, Milano / Virtual, Italy, Numéro 2020, 2021, Page(s) 4851-4856
Éditeur: IEEE
DOI: 10.1109/icpr48806.2021.9412884

Conversational Agents for Intelligent Buildings

Auteurs: Weronika Sieinska, Nancie Gunson, Christopher Walsh, Christian Dondrup, and Oliver Lemon
Publié dans: Proceedings of the SIGdial 2020 Conference, Numéro 2020, 2020, Page(s) 45-48
Éditeur: Association for Computational Linguistics

Why Approximate Matrix Square Root Outperforms Accurate SVD in Global Covariance Pooling?

Auteurs: Yue Song; Nicu Sebe; Wei Wang
Publié dans: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021
Éditeur: CVF
DOI: 10.48550/arxiv.2105.02498

Speech Modeling with a Hierarchical Transformer Dynamical VAE

Auteurs: Xiaoyu Lin; Xiaoyu Bie; Simon Leglaive; Laurent Girin; Xavier Alameda-Pineda
Publié dans: ICASSP 2023 - IEEE International Conference on Audio, Speech and Signal Processing, 2023
Éditeur: IEEE
DOI: 10.48550/arxiv.2303.09404

ODANet: Online Deep Appearance Network for Identity-Consistent Multi-person Tracking

Auteurs: Guillaume Delorme, Yutong Ban, Guillaume Sarrazin, Xavier Alameda-Pineda
Publié dans: ICPR 2021: Pattern Recognition. ICPR International Workshops and Challenges, 2021
Éditeur: Springer
DOI: 10.1007/978-3-030-68780-9_60

From two rolling shutters to one global shutter

Auteurs: Albl, Cenek; Kukelova, Zuzana; Larsson, Viktor; Pajdla, Tomas; Schindler, Konrad
Publié dans: CVPR, Numéro 2020, 2020
Éditeur: IEEE

Robust Relative Transfer Function Identification on Manifolds for Speech Enhancement

Auteurs: A. Sofer, T. Kounovský, J. Čmejla, Z. Koldovský and S. Gannot
Publié dans: 2021 29th European Signal Processing Conference (EUSIPCO), Numéro 2021, 2021, Page(s) 401-405
Éditeur: IEEE
DOI: 10.23919/eusipco54536.2021.9616175

Curriculum Graph Co-Teaching for Multi-Target Domain Adaptation

Auteurs: Subhankar Roy; Evgeny Krivosheev; Zhun Zhong; Nicu Sebe; Elisa Ricci
Publié dans: CVPR 2021, 2021, ISBN 978-1-6654-4509-2
Éditeur: Computer Vision Foundation
DOI: 10.5281/zenodo.5014029

Transformer-Based Attention Networks for Continuous Pixel-Wise Prediction

Auteurs: Guanglei Yang; Hao Tang; Mingli Ding; Nicu Sebe; Elisa Ricci
Publié dans: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Numéro 6, 2021
Éditeur: CVF
DOI: 10.1109/iccv48922.2021.01596

Robot control and navigation: ARI’s autonomous system

Auteurs: Francesco Ferro, Federico Nardi, Sara Cooper, Luca Marchionni
Publié dans: 29th IEEE International Conference on Robot & Human Interactive Communication ROMAN-2020, 2020
Éditeur: IEEE

Democratizing Fine-grained Visual Recognition with Large Language Models

Auteurs: Liu, Mingxuan; Roy, Subhankar; Li, Wenjing; Zhong, Zhun; Sebe, Nicu; Ricci, Elisa
Publié dans: ICLR 2024, 2024
Éditeur: ICLR
DOI: 10.48550/arxiv.2401.13837

Decoupled Direction-of-Arrival Estimations Using Relative Harmonic Coefficients

Auteurs: Y. Hu, T. D. Abhayapala, P. N. Samarasinghe and S. Gannot
Publié dans: 2020 28th European Signal Processing Conference (EUSIPCO), Numéro 2020, 2021, Page(s) 246-250
Éditeur: IEEE
DOI: 10.23919/eusipco47968.2020.9287611

A Conversational AI System for Tackling Misinformation

Auteurs: Nancie Gunson; Weronika Sieińska; Yanchao Yu; Daniel Hernandez Garcia; Jose L. Part; Christian Dondrup; Oliver Lemon
Publié dans: GoodIT '21: Proceedings of the Conference on Information Technology for Social Good, Numéro 2021, 2021, Page(s) 265-270
Éditeur: Association for Computing Machinery
DOI: 10.1145/3462203.3475874

Towards Visual Dialogue for Human-Robot Interaction

Auteurs: Jose L. Part; Daniel Hernández García; Yanchao Yu; Nancie Gunson; Christian Dondrup; Oliver Lemon
Publié dans: HRI '21 Companion: Companion of the 2021 ACM/IEEE International Conference on Human-Robot Interaction, Numéro 2021, 2021, Page(s) 670-672 (video submission)
Éditeur: IEEE
DOI: 10.1145/3434074.3447278

Smoothing the Disentangled Latent Style Space for Unsupervised Image-to-Image Translation

Auteurs: Yahui Liu; Sangineto, Enver; Yajing Chen; Linchao Bao; Haoxian Zhang; Sebe, Nicu; Lepri, Bruno; Wang, Wei; De Nadai, Marco
Publié dans: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 2021, 2021
Éditeur: IEEE
DOI: 10.5281/zenodo.5014014

D-InLoc++: Indoor Localization in Dynamic Environments

Auteurs: Martina Dubenova, Anna Zderadickova, Ondrej Kafka, Tomas Pajdla, Michal Polic
Publié dans: Lecture Notes in Computer Science book series (LNCS,volume 13485), 2022
Éditeur: Springer
DOI: 10.1007/978-3-031-16788-1_16

Recherche de données OpenAIRE...

Livrables

Publications

Partager cette page

Télécharger