Alexandre Alahi
EPFL ENAC IIC VITA
GC C1 383 (Bâtiment GC)
Station 18
1015 Lausanne
Web site: Web site: https://vita.epfl.ch/
EPFL ENAC IIC VITA
GC C1 383 (Bâtiment GC)
Station 18
1015 Lausanne
Web site: Web site: https://www.epfl.ch/schools/enac/about/diversity-office/
Fields of expertise
Mobility
Computer Vision
Machine Learning
Deep Learning
Human-Robot Interaction
Socially-aware Artificial Intelligence
Ambient Intelligence
Biography
Alexandre Alahi is an associate professor at EPFL leading the Visual Intelligence for Transportation laboratory (VITA). Before joining EPFL in 2017, he spent multiple years at Stanford University as a Post-doc and Research Scientist.His research lies at the intersection of Computer Vision, Machine Learning, and Robotics applied to transportation & mobility. To make Artificial Intelligence (AI) driven systems such as autonomous vehicles a safe reality, his lab works on a new type of Artificial Intelligence (AI), namely socially-aware AI, i.e., an AI augmented with social intelligence.
In 2022&2023, Alexandre was recognized as one of the top 100 Most Influential Scholar in Computer Vision over the past 10 years. His research team received the editor’s choice award from the journal Image and vision computing (2021) for their work on human motion prediction, the honorable mention at an ICCV workshop (2019) for their work on human pose estimation, the CVPR Open Source Award (2012) for their work on Retina-inspired image descriptors, and the ICDSC Challenge Prize (2009) for their sparsity-driven algorithm that has tracked more than 100 million pedestrians to date.
His work has been licensed to several companies and covered internationally by BBC, abc, PBS, Euronews, Wall street journal, and other national news outlets around the world. Alexandre has also co-founded multiple startups such as Visiosafe, and won several startup competitions. He was elected as one of the Top 20 Swiss Venture leaders in 2010.
Publications
Infoscience publications
Infoscience
TIC-TAC: A Framework for Improved Covariance Estimation in Deep Heteroscedastic Regression
2024-07-21. 41st International Conference on Machine Learning (ICML) 2024, Vienna, Austria, July 21-27, 2024.MaskCLR: Attention-Guided Contrastive Learning for Robust Action Representation Learning
2024-06-17. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, Washington, USA, June 17-21, 2024.Dynamic Voxels Based on Ego-Conditioned Prediction: An Integrated Spatio-Temporal Framework for Motion Planning
Ieee Transactions On Intelligent Transportation Systems. 2024-05-20. DOI : 10.1109/TITS.2024.3398008.Social-Transmotion: Promptable Human Trajectory Prediction
2024-05-01. International Conference on Learning Representations (ICLR), Vienna, Austria, May 7-11, 2024.When Your AI Becomes a Target: AI Security Incidents and Best Practices
2024-03-24. Proceedings of the 38th AAAI Conference on Artificial Intelligence, Vancouver, Canada, February 20-27, 2024. p. 23041-23046. DOI : 10.1609/aaai.v38i21.30347.Performing and Detecting Backdoor Attacks on Face Recognition Algorithms
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10656.Discrete choice modeling in the era of big data
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10456.Advancing Self-Supervised Deep Learning for 3D Scene Understanding
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10641.Spatial Modeling for Building Design Evaluation: from Visual Landscape Quality Assessment to Devaluation Risk Estimation
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10602.Toward Reliable Human Pose Forecasting With Uncertainty
IEEE Robotics and Automation Letters. 2024. DOI : 10.1109/LRA.2024.3374188.Predicting the long-term collective behaviour of fish pairs with deep learning
Journal of The Royal Society Interface. 2024. DOI : 10.1098/rsif.2023.0630.Safe Deep Neural Networks
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10384.Towards Motion Forecasting with Real-World Perception Inputs: Are End-to-End Approaches Competitive?
2024-05-15. International Conference on Robotics and Automation (ICRA), yokohama, japan, May 13- May 17, 2024.Smarter eco-cities and their leading-edge artificial intelligence of things solutions for environmental sustainability: A comprehensive systematic review
Environmental Science And Ecotechnology. 2023-11-03. DOI : 10.1016/j.ese.2023.100330.Toward contactless human thermal monitoring: A framework for Machine Learning-based human thermo-physiology modeling augmented with computer vision
Building and Environment. 2023-09-26. DOI : 10.1016/j.buildenv.2023.110850.Urban Digital Twin Challenges: A Systematic Review and Perspectives for Sustainable Smart Cities
Sustainable Cities And Society. 2023-08-29. DOI : 10.1016/j.scs.2023.104862.Real-Time Localization for Closed-Loop Control of Assistive Furniture
Ieee Robotics And Automation Letters. 2023-08-01. DOI : 10.1109/LRA.2023.3287365.Non-intrusive physiological parameters sensing for personalized human thermal comfort prediction
2023-06-11. 18th Healthy Buildings Europe Conference, Aachen, Germany, June 11-14, 2023.A generic diffusion-based approach for 3D human pose prediction in the wild
2023-05-29. IEEE International Conference on Robotics and Automation (ICRA), London, England, 29 May – 2 June, 2023. p. 8246-8253. DOI : 10.1109/ICRA48891.2023.10160399.The automated collapse data constructor technique and the data-driven methodology for seismic collapse risk assessment
Earthquake Engineering & Structural Dynamics. 2023-03-14. DOI : 10.1002/eqe.3865.Traffic perception from aerial images using butterfly fields
Transportation Research Part C-Emerging Technologies. 2023-08-01. DOI : 10.1016/j.trc.2023.104181.Manipulating Trajectory Prediction with Backdoors
2023Towards more Practical Threat Models in Artificial Intelligence Security
2023Making Computer Vision Models Robust and Adaptive
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9215.The inductive bias of deep learning: Connecting weights and functions
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9898.Building ventilation and outdoor air pollution: Implications for health, comfort and energy
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10558.Learning Robust and Adaptive Representations: from Interactions, for Interactions
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9863.Deep Generative Models for Autonomous Driving: from Motion Forecasting to Realistic Image Synthesis
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9876.Hardware-Software co-design Methodologies for Edge AI Optimization
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10318.Implicit Distance Functions: Learning and Applications in Robotics
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10197.Modeling Structured Data in Attention-based Models
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9804.Interpretable Representation Learning and Evaluation for Abstractive Summarization
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10045.Causal Triplet: An Open Challenge for Intervention-centric Causal Representation Learning
2023-04. Conference on Causal Learning and Reasoning (CLeaR), Tübingen, Germany, April 11-14, 2023.On Pitfalls of Test-time Adaptation
2023-7. International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA, July 23-29, 2023.Confidence Matters: Applications to Semantic Segmentation
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10055.Training Traffic Light Behavior with End-to-End Learning
2023. 17th International Conference on Intelligent Autonomous Systems (IAS), Zagreb, Croatia, June 13-16, 2022. p. 753-764. DOI : 10.1007/978-3-031-22216-0_50.Automated Human Motion Analysis and Synthesis
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-8976.Body Part-Based Representation Learning for Occluded Person Re-Identification
2023. Winter Conference on Applications of Computer Vision (WACV23), Waikoloa, Hawaii, January 3-7, 2023. p. 1613-1623. DOI : 10.1109/WACV56688.2023.00166.Safety-Compliant Generative Adversarial Networks for Human Trajectory Forecasting
IEEE Transactions on Intelligent Transportation Systems. 2023-01-18. DOI : 10.1109/TITS.2022.3233906.Composite Relationship Fields with Transformers for Scene Graph Generation
2023. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2023), Waikoloa, Hawaii, United States, January 3-7, 2023. p. 52-64. DOI : 10.1109/WACV56688.2023.00014.VL4Pose: Active Learning Through Out-Of-Distribution Detection For Pose Estimation
2022-11-21. 33rd British Machine Vision Conference (BMVC 2022), London, UK, November 21-24, 2022.Proxy methods for detection of inhalation exposure in simulated office environments
Journal Of Exposure Science And Environmental Epidemiology. 2022-11-08. DOI : 10.1038/s41370-022-00495-w.Are the trajectory prediction models generalizable?
2022-09-02. The IFAC Symposium on Robust Control Design (ROCOND), kyoto, japan, August 30-September 2, 2022.Vehicle trajectory prediction works, but not everywhere
2022-07-20. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2022), New Orleans, Louisiana, USA, June 21 - 24, 2022. DOI : 10.1109/CVPR52688.2022.01661.SoccerNet 2022 Challenges Results
2022-01-01. 5th ACM International Workshop on Multimedia Content Analysis in Sports (ACM MMSports), Lisboa, PORTUGAL, OCT 10-14, 2022. p. 75-86. DOI : 10.1145/3552437.3555703.STATNet: Spatial-temporal attention in the traffic prediction
2022-06-03. European Association for Research in Transportation (hEART), KU leuven, Belgium, June 1-3, 2022.Deep Learning Methods for Socially-Aware Human Trajectory Forecasting
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9842.Advancing the Adaptability of Compliant Robot Controllers for Exploration, Interaction, and Manipulation
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9747.Visual Scene Understanding for Transportation: From Detecting Objects To Relationships
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-8994.Unsupervised Visual Entity Abstraction towards 2D and 3D Compositional Models
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-8166.Motion Style Transfer: Modular Low-Rank Adaptation for Deep Motion Forecasting
2022. Conference on Robotics Learning, Auckland, New Zealand, December 14-18, 2022.Bridging the gap between model-driven and data-driven methods in the era of Big Data
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9112.Towards real-world biomechanical analysis of performance and functional capacity using wearable sensors
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9755.Efficient Transformer-Based Speech Recognition
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9766.Multivariate time series forecasting for freeway networks
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-8722.Towards Robust and Adaptive Motion Forecasting: A Causal Representation Perspective
2022. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2022), New Orleans, Louisiana, USA, June 21-24, 2022. p. 17060-17071. DOI : 10.1109/CVPR52688.2022.01657.Automated Detection Of Isolated Rem Sleep Behavior Disorder (Irbd) During Single Night In-Lab Video-Polysomnography (Psg) Using Computer Vision
2022-06-01. p. A282-A282.Motion Style TransferModular Low-Rank Style Transfer for Deep Trajectory Forecasting
2022-08-22Robustness and invariance properties of image classifiers
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9646.Pedestrian 3D Bounding Box Prediction
2022. 10th Symposium of the European Association for Research in Transportation (hEART), Leuven, Belgium, June, 2022.Are socially-aware trajectory prediction models really socially-aware?
Transportation Research Part C: Emerging Technologies. 2022. DOI : 10.1016/j.trc.2022.103705.Simple Yet Effective Action Recognition for Autonomous Driving
2022. 11th Triennial Symposium on Transportation Analysis Conference (TRISTAN XI), Mauritius Island, June 19-25, 2022.Theory and application of a data-driven approach to compressive spectrometry in the assessment of neurophotic stimulation
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9254.Stop Wasting my FLOPS: Improving the Efficiency of Deep Learning Models
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-8607.Learning to sample in Cartesian MRI
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9981.Reshaping Perception for Autonomous Driving with Semantic Keypoints
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-10072.Vision-based Flocking in Aerial Robot Swarms
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9186.Pedestrian Stop and Go Forecasting with Hybrid Feature Fusion
2022. IEEE 39th International Conference on Robotics and Automation (ICRA 2022), Philadelphia, Pennsylvania, USA, May 23-27, 2022. p. 940-947. DOI : 10.1109/ICRA46639.2022.9811664.Learning to Represent and Reconstruct 3D Deformable Objects
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-7974.OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association
Ieee Transactions On Intelligent Transportation Systems. 2022. DOI : 10.1109/TITS.2021.3124981.Detecting 32 Pedestrian Attributes for Autonomous Vehicles
IEEE Transactions on Intelligent Transportation Systems. 2022. DOI : 10.1109/TITS.2021.3107587.Perceiving Humans: from Monocular 3D Localization to Social Distancing
IEEE Transactions on Intelligent Transportation Systems. 2022. DOI : 10.1109/TITS.2021.3069376.A Shared Representation for Photorealistic Driving Simulators
IEEE Transactions on Intelligent Transportation Systems. 2021-12-03. DOI : 10.1109/TITS.2021.3131303.TTT++: When Does Self-Supervised Test-Time Training Fail or Thrive?
2021-12. 35th Annual Conference on Neural Information Processing Systems (NeurIPS 2022), Virtual, Decembre 6-14, 2021.Injecting Knowledge in Data-driven Vehicle Trajectory Predictors
Transportation Research Part C: Emerging Technologies. 2021-03-08. DOI : 10.1016/j.trc.2021.103010.Learning Decoupled Representations for Human Pose Forecasting
2021. 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCV 2021), Virtual, October 11-17, 2021. p. 2294-2303. DOI : 10.1109/ICCVW54120.2021.00259.DriverGym: Democratising Reinforcement Learning for Autonomous Driving
2021-11-12. NeurIPS 2021 Workshop on Machine Learning for Autonomous Driving, Online, December 6-14, 2021.Human Trajectory Forecasting in Crowds: A Deep Learning Perspective
IEEE Transactions on Intelligent Transportation Systems. 2021-04-19. DOI : 10.1109/TITS.2021.3069362.Efficient Depth-based Deep Learning Methods for Multi-Party Pose Estimation
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-8429.Gradient-based Methods for Deep Model Interpretability
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-8606.Damage Assessment of Stone Masonry Piers Using Imaged Surface Cracks
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-9252.Keypoint Communities
2021-10-11. IEEE International Conference on Computer Vision (ICCV 2021), [Virtual], October 10-17, 2021. p. 11037-11046. DOI : 10.1109/ICCV48922.2021.01087.Safety-aware Motion Prediction with Unseen Vehicles for Autonomous Driving
2021-10-11. IEEE International Conference on Computer Vision (ICCV 2021), [Virtual], October 10-17, 2021. p. 15731-15740. DOI : 10.1109/ICCV48922.2021.01544.Dynamic control strategies for managing pedestrian flows
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-8200.Architecture Exploration and Optimization of Heterogeneous Many-Core Compute and Memory Architectures with Architectural Extensions
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-7949.Social NCE: Contrastive Learning of Socially-aware Motion Representations
2021-10-11. IEEE International Conference on Computer Vision (ICCV), Virtual, October 10-17, 2021. p. 15098-15109. DOI : 10.1109/ICCV48922.2021.01484.Guest Editorial: Introduction to the Special Issue on Long-Term Human Motion Prediction
Ieee Robotics And Automation Letters. 2021-07-01. DOI : 10.1109/LRA.2021.3077964.Pedestrian intention prediction: A convolutional bottom-up multi-task approach
Transportation Research Part C: Emerging Technologies. 2021-07-15. DOI : 10.1016/j.trc.2021.103259.Interpretable Social Anchors for Human Trajectory Forecasting in Crowds
2021. Conference on Computer Vision and Pattern Recognition (CVPR), 2021, Nashville, USA (Virtual), Jun 19–25, 2021. p. 15551-15561. DOI : 10.1109/CVPR46437.2021.01530.Deep visual Re-identification with confidence
Transportation Research Part C: Emerging Technologies. 2021. DOI : 10.1016/j.trc.2021.103067.MonStereo: When Monocular and Stereo Meet at the Tail of 3D Human Localization
2021-03-22. International Conference on Robotics and Automation (ICRA), Xi'an, China, June 1-6, 2021. p. 5126-5132. DOI : 10.1109/ICRA48506.2021.9561820.Adaptive hot water production based on Supervised Learning
Sustainable Cities and Society. 2021. DOI : 10.1016/j.scs.2020.102625.Detecting 32 Pedestrian Attributes for Autonomous Vehicles
2020-12-04Gait in real world: validated algorithms for gait periods and speed estimation using a single wearable sensor
Lausanne, EPFL, 2020. DOI : 10.5075/epfl-thesis-8099.Automatic detection of hand hygiene using computer vision technology
Journal of the American Medical Informatics Association. 2020. DOI : 10.1093/jamia/ocaa115.Pedestrian Intention Prediction: A Multi-task Perspective
2020. 9th Symposium of the European Association for Research in Transportation (hEART 2021), Lyon, France, 7-9 April 2021.Next Steps for Image Synthesis using Semantic Segmentation
2020. 20th Swiss Transport Research Conference (STRC 2020), [Online event], 13-14 May 2020.DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation
2020-09-29. NeurIPS 2020 34th Conference on Neural Information Processing Systems, Vancouver, Canada, December 6-12, 2020.The electric autonomous dial-a-ride problem
Lausanne, EPFL, 2020. DOI : 10.5075/epfl-thesis-7431.Precise Hand Finger Width Estimation via RGB-D Data
2020-08-31TIC-TAC: A Framework for Improved Covariance Estimation in Deep Heteroscedastic Regression
2024-07-21. 41st International Conference on Machine Learning (ICML) 2024, Vienna, Austria, July 21-27, 2024.MaskCLR: Attention-Guided Contrastive Learning for Robust Action Representation Learning
2024-06-17. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, Washington, USA, June 17-21, 2024.Dynamic Voxels Based on Ego-Conditioned Prediction: An Integrated Spatio-Temporal Framework for Motion Planning
Ieee Transactions On Intelligent Transportation Systems. 2024-05-20. DOI : 10.1109/TITS.2024.3398008.Social-Transmotion: Promptable Human Trajectory Prediction
2024-05-01. International Conference on Learning Representations (ICLR), Vienna, Austria, May 7-11, 2024.When Your AI Becomes a Target: AI Security Incidents and Best Practices
2024-03-24. Proceedings of the 38th AAAI Conference on Artificial Intelligence, Vancouver, Canada, February 20-27, 2024. p. 23041-23046. DOI : 10.1609/aaai.v38i21.30347.Performing and Detecting Backdoor Attacks on Face Recognition Algorithms
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10656.Discrete choice modeling in the era of big data
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10456.Advancing Self-Supervised Deep Learning for 3D Scene Understanding
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10641.Spatial Modeling for Building Design Evaluation: from Visual Landscape Quality Assessment to Devaluation Risk Estimation
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10602.Toward Reliable Human Pose Forecasting With Uncertainty
IEEE Robotics and Automation Letters. 2024. DOI : 10.1109/LRA.2024.3374188.Predicting the long-term collective behaviour of fish pairs with deep learning
Journal of The Royal Society Interface. 2024. DOI : 10.1098/rsif.2023.0630.Safe Deep Neural Networks
Lausanne, EPFL, 2024. DOI : 10.5075/epfl-thesis-10384.Towards Motion Forecasting with Real-World Perception Inputs: Are End-to-End Approaches Competitive?
2024-05-15. International Conference on Robotics and Automation (ICRA), yokohama, japan, May 13- May 17, 2024.Smarter eco-cities and their leading-edge artificial intelligence of things solutions for environmental sustainability: A comprehensive systematic review
Environmental Science And Ecotechnology. 2023-11-03. DOI : 10.1016/j.ese.2023.100330.Toward contactless human thermal monitoring: A framework for Machine Learning-based human thermo-physiology modeling augmented with computer vision
Building and Environment. 2023-09-26. DOI : 10.1016/j.buildenv.2023.110850.Urban Digital Twin Challenges: A Systematic Review and Perspectives for Sustainable Smart Cities
Sustainable Cities And Society. 2023-08-29. DOI : 10.1016/j.scs.2023.104862.Real-Time Localization for Closed-Loop Control of Assistive Furniture
Ieee Robotics And Automation Letters. 2023-08-01. DOI : 10.1109/LRA.2023.3287365.Non-intrusive physiological parameters sensing for personalized human thermal comfort prediction
2023-06-11. 18th Healthy Buildings Europe Conference, Aachen, Germany, June 11-14, 2023.A generic diffusion-based approach for 3D human pose prediction in the wild
2023-05-29. IEEE International Conference on Robotics and Automation (ICRA), London, England, 29 May – 2 June, 2023. p. 8246-8253. DOI : 10.1109/ICRA48891.2023.10160399.The automated collapse data constructor technique and the data-driven methodology for seismic collapse risk assessment
Earthquake Engineering & Structural Dynamics. 2023-03-14. DOI : 10.1002/eqe.3865.Traffic perception from aerial images using butterfly fields
Transportation Research Part C-Emerging Technologies. 2023-08-01. DOI : 10.1016/j.trc.2023.104181.Manipulating Trajectory Prediction with Backdoors
2023Towards more Practical Threat Models in Artificial Intelligence Security
2023Making Computer Vision Models Robust and Adaptive
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9215.The inductive bias of deep learning: Connecting weights and functions
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9898.Building ventilation and outdoor air pollution: Implications for health, comfort and energy
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10558.Learning Robust and Adaptive Representations: from Interactions, for Interactions
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9863.Deep Generative Models for Autonomous Driving: from Motion Forecasting to Realistic Image Synthesis
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9876.Hardware-Software co-design Methodologies for Edge AI Optimization
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10318.Implicit Distance Functions: Learning and Applications in Robotics
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10197.Modeling Structured Data in Attention-based Models
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9804.Interpretable Representation Learning and Evaluation for Abstractive Summarization
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10045.Causal Triplet: An Open Challenge for Intervention-centric Causal Representation Learning
2023-04. Conference on Causal Learning and Reasoning (CLeaR), Tübingen, Germany, April 11-14, 2023.On Pitfalls of Test-time Adaptation
2023-7. International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA, July 23-29, 2023.Confidence Matters: Applications to Semantic Segmentation
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-10055.Training Traffic Light Behavior with End-to-End Learning
2023. 17th International Conference on Intelligent Autonomous Systems (IAS), Zagreb, Croatia, June 13-16, 2022. p. 753-764. DOI : 10.1007/978-3-031-22216-0_50.Automated Human Motion Analysis and Synthesis
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-8976.Body Part-Based Representation Learning for Occluded Person Re-Identification
2023. Winter Conference on Applications of Computer Vision (WACV23), Waikoloa, Hawaii, January 3-7, 2023. p. 1613-1623. DOI : 10.1109/WACV56688.2023.00166.Safety-Compliant Generative Adversarial Networks for Human Trajectory Forecasting
IEEE Transactions on Intelligent Transportation Systems. 2023-01-18. DOI : 10.1109/TITS.2022.3233906.Composite Relationship Fields with Transformers for Scene Graph Generation
2023. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2023), Waikoloa, Hawaii, United States, January 3-7, 2023. p. 52-64. DOI : 10.1109/WACV56688.2023.00014.VL4Pose: Active Learning Through Out-Of-Distribution Detection For Pose Estimation
2022-11-21. 33rd British Machine Vision Conference (BMVC 2022), London, UK, November 21-24, 2022.Proxy methods for detection of inhalation exposure in simulated office environments
Journal Of Exposure Science And Environmental Epidemiology. 2022-11-08. DOI : 10.1038/s41370-022-00495-w.Are the trajectory prediction models generalizable?
2022-09-02. The IFAC Symposium on Robust Control Design (ROCOND), kyoto, japan, August 30-September 2, 2022.Vehicle trajectory prediction works, but not everywhere
2022-07-20. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2022), New Orleans, Louisiana, USA, June 21 - 24, 2022. DOI : 10.1109/CVPR52688.2022.01661.SoccerNet 2022 Challenges Results
2022-01-01. 5th ACM International Workshop on Multimedia Content Analysis in Sports (ACM MMSports), Lisboa, PORTUGAL, OCT 10-14, 2022. p. 75-86. DOI : 10.1145/3552437.3555703.STATNet: Spatial-temporal attention in the traffic prediction
2022-06-03. European Association for Research in Transportation (hEART), KU leuven, Belgium, June 1-3, 2022.Deep Learning Methods for Socially-Aware Human Trajectory Forecasting
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9842.Advancing the Adaptability of Compliant Robot Controllers for Exploration, Interaction, and Manipulation
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9747.Visual Scene Understanding for Transportation: From Detecting Objects To Relationships
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-8994.Unsupervised Visual Entity Abstraction towards 2D and 3D Compositional Models
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-8166.Motion Style Transfer: Modular Low-Rank Adaptation for Deep Motion Forecasting
2022. Conference on Robotics Learning, Auckland, New Zealand, December 14-18, 2022.Bridging the gap between model-driven and data-driven methods in the era of Big Data
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9112.Towards real-world biomechanical analysis of performance and functional capacity using wearable sensors
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9755.Efficient Transformer-Based Speech Recognition
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9766.Multivariate time series forecasting for freeway networks
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-8722.Towards Robust and Adaptive Motion Forecasting: A Causal Representation Perspective
2022. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2022), New Orleans, Louisiana, USA, June 21-24, 2022. p. 17060-17071. DOI : 10.1109/CVPR52688.2022.01657.Automated Detection Of Isolated Rem Sleep Behavior Disorder (Irbd) During Single Night In-Lab Video-Polysomnography (Psg) Using Computer Vision
2022-06-01. p. A282-A282.Motion Style TransferModular Low-Rank Style Transfer for Deep Trajectory Forecasting
2022-08-22Robustness and invariance properties of image classifiers
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9646.Pedestrian 3D Bounding Box Prediction
2022. 10th Symposium of the European Association for Research in Transportation (hEART), Leuven, Belgium, June, 2022.Are socially-aware trajectory prediction models really socially-aware?
Transportation Research Part C: Emerging Technologies. 2022. DOI : 10.1016/j.trc.2022.103705.Simple Yet Effective Action Recognition for Autonomous Driving
2022. 11th Triennial Symposium on Transportation Analysis Conference (TRISTAN XI), Mauritius Island, June 19-25, 2022.Theory and application of a data-driven approach to compressive spectrometry in the assessment of neurophotic stimulation
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9254.Stop Wasting my FLOPS: Improving the Efficiency of Deep Learning Models
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-8607.Learning to sample in Cartesian MRI
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9981.Reshaping Perception for Autonomous Driving with Semantic Keypoints
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-10072.Vision-based Flocking in Aerial Robot Swarms
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9186.Pedestrian Stop and Go Forecasting with Hybrid Feature Fusion
2022. IEEE 39th International Conference on Robotics and Automation (ICRA 2022), Philadelphia, Pennsylvania, USA, May 23-27, 2022. p. 940-947. DOI : 10.1109/ICRA46639.2022.9811664.Learning to Represent and Reconstruct 3D Deformable Objects
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-7974.OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association
Ieee Transactions On Intelligent Transportation Systems. 2022. DOI : 10.1109/TITS.2021.3124981.Detecting 32 Pedestrian Attributes for Autonomous Vehicles
IEEE Transactions on Intelligent Transportation Systems. 2022. DOI : 10.1109/TITS.2021.3107587.Perceiving Humans: from Monocular 3D Localization to Social Distancing
IEEE Transactions on Intelligent Transportation Systems. 2022. DOI : 10.1109/TITS.2021.3069376.A Shared Representation for Photorealistic Driving Simulators
IEEE Transactions on Intelligent Transportation Systems. 2021-12-03. DOI : 10.1109/TITS.2021.3131303.TTT++: When Does Self-Supervised Test-Time Training Fail or Thrive?
2021-12. 35th Annual Conference on Neural Information Processing Systems (NeurIPS 2022), Virtual, Decembre 6-14, 2021.Injecting Knowledge in Data-driven Vehicle Trajectory Predictors
Transportation Research Part C: Emerging Technologies. 2021-03-08. DOI : 10.1016/j.trc.2021.103010.Learning Decoupled Representations for Human Pose Forecasting
2021. 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCV 2021), Virtual, October 11-17, 2021. p. 2294-2303. DOI : 10.1109/ICCVW54120.2021.00259.DriverGym: Democratising Reinforcement Learning for Autonomous Driving
2021-11-12. NeurIPS 2021 Workshop on Machine Learning for Autonomous Driving, Online, December 6-14, 2021.Human Trajectory Forecasting in Crowds: A Deep Learning Perspective
IEEE Transactions on Intelligent Transportation Systems. 2021-04-19. DOI : 10.1109/TITS.2021.3069362.Efficient Depth-based Deep Learning Methods for Multi-Party Pose Estimation
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-8429.Gradient-based Methods for Deep Model Interpretability
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-8606.Damage Assessment of Stone Masonry Piers Using Imaged Surface Cracks
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-9252.Keypoint Communities
2021-10-11. IEEE International Conference on Computer Vision (ICCV 2021), [Virtual], October 10-17, 2021. p. 11037-11046. DOI : 10.1109/ICCV48922.2021.01087.Safety-aware Motion Prediction with Unseen Vehicles for Autonomous Driving
2021-10-11. IEEE International Conference on Computer Vision (ICCV 2021), [Virtual], October 10-17, 2021. p. 15731-15740. DOI : 10.1109/ICCV48922.2021.01544.Dynamic control strategies for managing pedestrian flows
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-8200.Architecture Exploration and Optimization of Heterogeneous Many-Core Compute and Memory Architectures with Architectural Extensions
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-7949.Social NCE: Contrastive Learning of Socially-aware Motion Representations
2021-10-11. IEEE International Conference on Computer Vision (ICCV), Virtual, October 10-17, 2021. p. 15098-15109. DOI : 10.1109/ICCV48922.2021.01484.Guest Editorial: Introduction to the Special Issue on Long-Term Human Motion Prediction
Ieee Robotics And Automation Letters. 2021-07-01. DOI : 10.1109/LRA.2021.3077964.Pedestrian intention prediction: A convolutional bottom-up multi-task approach
Transportation Research Part C: Emerging Technologies. 2021-07-15. DOI : 10.1016/j.trc.2021.103259.Interpretable Social Anchors for Human Trajectory Forecasting in Crowds
2021. Conference on Computer Vision and Pattern Recognition (CVPR), 2021, Nashville, USA (Virtual), Jun 19–25, 2021. p. 15551-15561. DOI : 10.1109/CVPR46437.2021.01530.Deep visual Re-identification with confidence
Transportation Research Part C: Emerging Technologies. 2021. DOI : 10.1016/j.trc.2021.103067.MonStereo: When Monocular and Stereo Meet at the Tail of 3D Human Localization
2021-03-22. International Conference on Robotics and Automation (ICRA), Xi'an, China, June 1-6, 2021. p. 5126-5132. DOI : 10.1109/ICRA48506.2021.9561820.Adaptive hot water production based on Supervised Learning
Sustainable Cities and Society. 2021. DOI : 10.1016/j.scs.2020.102625.Detecting 32 Pedestrian Attributes for Autonomous Vehicles
2020-12-04Gait in real world: validated algorithms for gait periods and speed estimation using a single wearable sensor
Lausanne, EPFL, 2020. DOI : 10.5075/epfl-thesis-8099.Automatic detection of hand hygiene using computer vision technology
Journal of the American Medical Informatics Association. 2020. DOI : 10.1093/jamia/ocaa115.Pedestrian Intention Prediction: A Multi-task Perspective
2020. 9th Symposium of the European Association for Research in Transportation (hEART 2021), Lyon, France, 7-9 April 2021.Next Steps for Image Synthesis using Semantic Segmentation
2020. 20th Swiss Transport Research Conference (STRC 2020), [Online event], 13-14 May 2020.DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation
2020-09-29. NeurIPS 2020 34th Conference on Neural Information Processing Systems, Vancouver, Canada, December 6-12, 2020.The electric autonomous dial-a-ride problem
Lausanne, EPFL, 2020. DOI : 10.5075/epfl-thesis-7431.Precise Hand Finger Width Estimation via RGB-D Data
2020-08-31Research
Research summary
We work on the theoretical challenges and practical applications of socially-aware systems, i.e., machines that can not only perceive human behavior, but reason with social intelligence in the context of transportation problems and smart spaces. We envision a future where intelligent machines are ubiquitous, where self-driving cars, delivery robots, and self-moving Segways are facts of everyday life. Beyond embodied agents, we will also see our living spaces – our homes, buildings, and cities – become equipped with ambient intelligence which can sense and respond to human behavior. However, to realize this future, intelligent machines need to develop social intelligence and the ability to make safe and consistent decisions in unconstrained crowded social scenes. Self-driving vehicles must learn social etiquette in order to navigate cities like Paris or Naples. Social robots need to comply with social conventions and obey (unwritten) common-sense rules to effectively operate in crowded terminals. For instance, they need to respect personal space, yield right-of-way, and “read” the behavior of others to predict future actions. Our research is centered around understanding and predicting human social behavior with multi-modal visual data. Our work spans multiple aspects of socially-aware systems: from 1- collecting multi-modal data at scale, 2- Extracting coarse-to-fine grained behaviours in real-time, 3- designing deep learning methods that can learn to predict human social behavior in a fully data-driven way, to 4- integrating the developed methods in real-world systems such as a vehicle or a socially-aware robot that navigates crowded social scenes.Teaching & PhD
Teaching
Civil Engineering