CHEPFLCivil Engineering Institute (IIC, School of Architecture, Civil and Environmental Engineering ENAC)
Alexandre Alahi
Associate Professor
Leads the Visual Intelligence for Transportation lab, developing computer vision and deep learning for autonomous mobility, such as human trajectory forecasting and multi-agent perception.
Alexandre Alahi's lab at EPFL works on embodied & active vision, robot perception & SLAM and time series, anomaly & spatio-temporal data. 107 papers since 2023, 45 at top venues, mostly CVPR and NeurIPS. 12 PhD students. 8 alumni, most often to Google. Applications close .
At a glance
- papers, 2023–2026
- 28/ year
- at T1 venues
- 42%
- PhD students
- 12
- MS students
- 0
- alumni listed
- 8
Papers by yearfrom the lab homepage, 2023–2026
- 24 papers2026 · T1 12 · collecting
- 37 papers2025 · T1 23
- 24 papers2024 · T1 8
- 22 papers2023 · T1 2
T1 venuesOtherVenues: CVPR, NeurIPS, ICLR, ECCV, ICML
Research focuspaper titles that use each topic's words
- Time series, anomaly & spatio-temporal data11 papers
- Embodied & active vision8 papers
- Robot perception & SLAM2 papers
- civil0 papers
- visual intelligence0 papers
Paper list
From the lab homepage.
202624 papers
- SLEEPJ1172 Training Artificial Intelligence Models to Assess Injury Risk in REM Sleep Behavior Disorder (opens in a new tab)
- APOS Trends in OrthodonticsComputer-assisted development of age- and sex-specific cephalometric templates for evaluating facial profiles: A cross-sectional study (opens in a new tab)
- CVPR (T1)Deformable Gaussian Occupancy: Decoupling Rigid and Nonrigid Motion with Factorized Distillation
- ACM Conference on User Modeling, Adaptation and PersonalizationEvaluating Persona Consistency and Emotion Understanding in Small Language Models (opens in a new tab)
- NeurIPS (T1)EverAnimate: Minute-Scale Human Animation via Latent Flow Restoration
- Scientific ReportsExplainable AI–Enhanced Pavement Deterioration Modeling Using Hybrid Classification–Regression Deep Learning Architecture (opens in a new tab)
- Mobility Patterns, Big Data and Transport Analytics: Tools and Applications for ModelingFoundational principles for representation learning: applications in autonomous mobility (opens in a new tab)
- IEEE International Conference on Robotics & AutomationFusing Satellite Imagery and Planimetric Maps for Cross-View Localization
- ICLR (T1)LayerSync: Self-aligning Intermediate Layers
- ICLR (T1)Loc2: Interpretable Cross-View Localization via Depth-Lifted Local Feature Matching
- Physical Review LettersObservation of a Threshold Enhancement in the π+π- Spectrum in ψ (3686) →π+π-J/ψ Decays (opens in a new tab)
- Physical Review LettersObservation of the Electromagnetic Radiative Decays of the Λ(1520) and Λ(1690) to γΣ0 (opens in a new tab)
202537 papers
- ECCV (T1)Adapting Fine-Grained Cross-View Localization to Areas Without Fine Ground Truth (opens in a new tab)
- Annals of NeurologyAutomated Detection of Isolated REM Sleep Behavior Disorder Using Computer Vision (opens in a new tab)
- IROS (T1)Boosting Omnidirectional Stereo Matching with a Pre-trained Depth Foundation Model (opens in a new tab)
- CVPR (T1)Certified Human Trajectory Prediction (opens in a new tab)
- IROS (T1)COARSE: Collaborative Pseudo-Labeling with Coarse Real Labels for Off-Road Semantic Segmentation (opens in a new tab)
- CVPR (T1)CoMatcher: Multi-View Collaborative Feature Matching
- Case Studies In Construction MaterialsData generation for asphalt pavement evaluation: Deep learning-based insights from generative models (opens in a new tab)
- NeurIPS (T1)EvoLM: In Search of Lost Training Dynamics for Language Model Reasoning (opens in a new tab)
- CVPR (T1)FG2 : Fine-Grained Cross-View Localization by Fine-Grained Feature Matching (opens in a new tab)
- CVPR (T1)GEM: A Generalizable Ego-Vision Multimodal World Model for Fine-Grained Ego-Motion, Object Dynamics, and Scene Composition Control (opens in a new tab)
- ICCV (T1)GeoDistill: Geometry-Guided Self-Distillation for Weakly Supervised Cross-View Localization
- CVPR (T1)HELVIPAD: A Real-World Dataset for Omnidirectional Stereo Depth Estimation (opens in a new tab)
202424 papers
- Transportation Research Part C: Emerging TechnologiesA qualitative AI security risk assessment of autonomous vehicles (opens in a new tab)
- The Annual Meeting of the Associated Professional Sleep SocietiesAutomated Detection of Isolated REM Sleep Behaviour Disorder Using Computer Vision (opens in a new tab)
- International Journal of Machine Learning and CyberneticsBackdoor learning curves: explaining backdoor poisoning beyond influence functions (opens in a new tab)
- —CODE: Confident Ordinary Differential Editing (opens in a new tab)
- IEEE Transactions on Intelligent VehiclesCrossFeat: Semantic Cross-modal Attention for Pedestrian Behavior Forecasting (opens in a new tab)
- IEEE Transactions on Intelligent Transportation SystemsDynamic Voxels Based on Ego-Conditioned Prediction: An Integrated Spatio-Temporal Framework for Motion Planning (opens in a new tab)
- Translational vision science & technologyEye-Rubbing Detection Using a Smartwatch: A Feasibility Study Demonstrated High Accuracy With Machine Learning (opens in a new tab)
- STRCJRDB-Traj: A Dataset and Benchmark for Trajectory Forecasting in Crowds
- ComputerMachine Learning Security Against Data Poisoning: Are We There Yet? (opens in a new tab)
- CVPR (T1)MaskCLR: Attention-Guided Contrastive Learning for Robust Action Representation Learning
- T-CAP ECCV24MPL: Lifting 3D Human Pose from Multi-view 2D Poses (opens in a new tab)
- CoRL (T1)Multi-Transmotion: Pre-trained Model for Human Motion Prediction
202322 papers
- ICRA (T1)A generic diffusion-based approach for 3D human pose prediction in the wild (opens in a new tab)
- Machine LearningAdversarial vulnerability bounds for Gaussian process classification (opens in a new tab)
- WACVBody Part-Based Representation Learning for Occluded Person Re-Identification (opens in a new tab)
- CLeaRCausal Triplet: An Open Challenge for Intervention-centric Causal Representation Learning
- WACVComposite Relationship Fields with Transformers for Scene Graph Generation (opens in a new tab)
- MMSportsDeepSportradar-v2: a Multi-Sport Computer Vision Dataset for Sport Understandings (opens in a new tab)
- IEEE Transactions on Information Forensics and SecurityMachine Learning Security in Industry: A Quantitative Survey (opens in a new tab)
- —Manipulating Trajectory Prediction with Backdoors
- MMSportsMulti-task Learning for Joint Re-identification, Team Affiliation, and Role Classification for Sports Visual Tracking (opens in a new tab)
- Healthy Buildings Europe ConferenceNon-intrusive physiological parameters sensing for personalized human thermal comfort prediction
- ICML (T1)On Pitfalls of Test-time Adaptation
- RA-LReal-Time Localization for Closed-Loop Control of Assistive Furniture (opens in a new tab)
Where alumni went8 listed, 25% to industry
Industry 1Academia 3
Most often to
Also to
- EPFL
- Stanford University
Apply to EPFLFall 2027
EDIC Computer and Communication Sciences PhD
- Round
- Round 1
- Opens
- —
- GRE
- optional
- English
- TOEFL 100 / IELTS 7.0
- Fee
- none
Confirmed on the program page.
Similar labsnearest by research summary
- CNSJTU
Jianhua Sun
- Multimodal & grounded language
- SGNUS
Wee Kheng Leow
- 3D vision & neural rendering
- CNZhejiang
Yi Yang
- Embodied & active vision
- CNSJTU
Yong-Lu Li
- Reasoning & agents
- CNTsinghua
Jiwen Lu
- Embodied & active vision
- GBEdinburgh
Oisin Mac Aodha
- 3D vision & neural rendering
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
EPFLcloses