Fields of expertise
BiographyPascal Fua received an engineering degree from Ecole Polytechnique, Paris, in 1984 and the Ph.D. degree in Computer Science from the University of Orsay in 1989. He then worked at SRI International and INRIA Sophia-Antipolis as a Computer Scientist. He joined EPFL in 1996 where he is now a Professor in the School of Computer and Communication Science and heads the Computer Vision Laboratory.
His research interests include shape modeling and motion recovery from images, analysis of microscopy images, and Augmented Reality. His research interests include shape modeling and motion recovery from images, analysis of microscopy images, and machine learning. He has (co)authored over 300 publications in refereed journals and conferences. He is an IEEE Fellow and has been an Associate Editor of IEEE journal Transactions for Pattern Analysis and Machine Intelligence. He often serves as program committee member, area chair, and program chair of major vision conferences and has cofounded three spinoff companies (Pix4D, PlayfulVision, and NeuralConcept).
Current WorkThe research activities of the Computer Vision Laboratory focus on shape and motion recovery from images, object and people detection and tracking in video sequences, and analysis of brain microscopy image-stacks. CVLab also provides undergraduate and graduate teaching and performs technology transfer to both established and start up companies.
Pascal Fua's research has been sponsored by the Swiss National Science Foundation, Innosuisse, the European Union including a senior ERC grant, and several industrial partners.
All since 1996
Multi-view Tracking Using Weakly Supervised Human Motion Prediction2023. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2023), Waikoloa, Hawaii, USA, January 3-7, 2023.
Two-level Data Augmentation for Calibrated Multi-view Detection2023. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2023), Waikoloa, Hawaii, USA, January 3-7, 2023.
3D Pose Based Feedback For Physical Exercises2022-11-24. 16th Asian Conference on Computer Vision (ACCV 2022), Macau, China, December 4-8, 2022. p. 1316-1332.
Weakly Supervised Volumetric Image Segmentation with Deformed Templates2022-09-18. 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2022), Singapore, September 18-22, 2022. p. 422-432. DOI : 10.1007/978-3-031-16443-9_41.
Long Term Motion Prediction Using Keyposes2022-09-12. 3DV 2022, Prague, Czechia, September 12-15, 2022.
Generating LOD3 building models from structure-from-motion and semantic segmentationAutomation In Construction. 2022-09-01. DOI : 10.1016/j.autcon.2022.104430.
TOPO-Loss for continuity-preserving crack detection using deep learningConstruction And Building Materials. 2022-08-15. DOI : 10.1016/j.conbuildmat.2022.128264.
Improving the Training of Compact Neural Networks for Visual RecognitionLausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9061.
Latent Representation of CFD Meshes and Application to 2D Airfoil Aerodynamics2022
Neural Annotation Refinement: Development of a New 3D Dataset for Adrenal Gland Analysis2022-01-01. 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Singapore, SINGAPORE, Sep 18-22, 2022. p. 503-513. DOI : 10.1007/978-3-031-16440-8_48.
Enforcing Connectivity of 3D Linear Structures Using Their 2D Projections2022-01-01. 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Singapore, SINGAPORE, Sep 18-22, 2022. p. 591-601. DOI : 10.1007/978-3-031-16443-9_57.
Learning to Simulate Realistic LiDARs2022. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2022), Kyoto, Japan, October 23-27, 2022.
DIG: Draping Implicit Garment over the Human Body2022. Asian Conference on Computer Vision (ACCV 2022), Macau SAR, China, December 4-8, 2022.
Understanding Deep Neural Networks using Adversarial AttacksLausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9259.
MeshUDF: Fast and Differentiable Meshing of Unsigned Distance Field Networks2022. European Conference on Computer Vision (ECCV 2022), Tel-Aviv, Israel, October 23-27, 2022. p. 14.
Learning to Align Sequential Actions in the Wild2022. CVPR 2022 : IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans , United States, June 21 - 24, 2022.
Structure-aware Multi-view 3D Reconstruction of Dislocations in TEM with Message Passing Neural NetworksJoint Meeting of Dreiländertagung & Multinational Congress on Microscopy (Microscopy Conference 2021), Digital, August 22-26, 2021.
Eigendecomposition-Free Training of Deep Networks for Linear Least-Square ProblemsIeee Transactions On Pattern Analysis And Machine Intelligence. 2021-09-01. DOI : 10.1109/TPAMI.2020.2978812.
LiftPose3D, a deep learning-based approach for transforming two-dimensional to three-dimensional poses in laboratory animalsNature Methods. 2021-08-01. DOI : 10.1038/s41592-021-01226-z.
Drainage Canals in Southeast Asian Peatlands Increase Carbon EmissionsAgu Advances. 2021-03-01. DOI : 10.1029/2020AV000321.
Leveraging Spatial and Photometric Context for Calibrated Non-Lambertian Photometric Stereo2021-01-01. 9th International Conference on 3D Vision (3DV), ELECTR NETWORK, Dec 01-03, 2021. p. 394-402. DOI : 10.1109/3DV53792.2021.00049.
Image Matching Across Wide Baselines: From Paper to PracticeInternational Journal Of Computer Vision. 2021. DOI : 10.1007/s11263-020-01385-0.
TopoAL: An Adversarial Learning Approach for Topology-Aware Road Segmentation2020-08-23. European Conference on Computer Vision (ECCV), [Online event], August 23-28, 2020.
XNect: Real-time Multi-Person 3D Motion Capture with a Single RGB CameraAcm Transactions On Graphics (TOG). 2020-07-01. DOI : 10.1145/3386569.3392410.
DISK: learning local features with policy gradient2020-06-01. Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada (online), December 6-12, 2020.
MeshSDF: Differentiable Iso-Surface Extraction2020-06-01. Conference on Neural Information Processing Systems (NeurIPS), [Virtual event], December, 2020.
Better Patch Stitching for Parametric Surface Reconstruction2020-01-01. 8th International Conference on 3D Vision (3DV), ELECTR NETWORK, Nov 25-28, 2020. p. 593-602. DOI : 10.1109/3DV50981.2020.00069.
Local Non-Rigid Structure-from-Motion from Diffeomorphic Mappings2020-01-01. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), ELECTR NETWORK, Jun 14-19, 2020. p. 2056-2064. DOI : 10.1109/CVPR42600.2020.00213.
UCLID-Net: Single View Reconstruction in Object Space2020. 34th Conference on Neural Information Processing Systems, Virtual, December 6-12, 2020.
Towards Reliable Evaluation of Algorithms for Road Network Reconstruction from Aerial Images2020. European Conference on Computer Vision (ECCV 2020), [Virtual conference], August 23-28, 2020.
Voxel2Mesh: 3D Mesh Model Generation from Volumetric Data2020. International Conference On Medical Image Computing & Computer Assisted Intervention (MICCAI), Lima, Peru, 4-8 OCTOBER 2020.
Local Non-Rigid Structure-from-Motion from Locally Diffeomorphic Mappings2020. Computer Vision and Pattern Recognition (CVPR), Seattle, USA, June 16-18, 2020.
Visual Correspondences for Unsupervised Domain Adaptation on Electron Microscopy ImagesIEEE Transactions on Medical Imaging. 2020. DOI : 10.1109/TMI.2019.2946462.
Domain-Adaptive Multibranch Networks2020. International Conference on Learning Representations (ICLR), Addis Ababa, Ethiopia, April 26-30, 2020.
Aerodynamic shape optimization via surrogate modelling with convolutional neural networks2019-06-21
Geometric Deep Learning for Volumetric Computational Fluid Dynamics2019-06-21
Mo(2)Cap(2): Real-time Mobile 3D Motion Capture with a Cap-mounted Fisheye CameraIEEE Transactions On Visualization And Computer Graphics. 2019-05-01. DOI : 10.1109/TVCG.2019.2898650.
Geometry in active learning for binary and multi-class image segmentationComputer Vision And Image Understanding (CVIU). 2019-05-01. DOI : 10.1016/j.cviu.2019.01.007.
What Face and Body Shapes Can Tell Us About Height2019-01-01. IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, SOUTH KOREA, Oct 27-Nov 02, 2019. p. 1819-1827. DOI : 10.1109/ICCVW.2019.00226.
Beyond Cartesian Representations for Local Descriptors2019-01-01. IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, SOUTH KOREA, Oct 27-Nov 02, 2019. p. 253-262. DOI : 10.1109/ICCV.2019.00034.
Gravity as a Reference for Estimating a Person's Height from Video2019-01-01. IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, SOUTH KOREA, Oct 27-Nov 02, 2019. p. 8568-8576. DOI : 10.1109/ICCV.2019.00866.
GarNet: A Two-Stream Network for Fast and Accurate 3D Cloth Draping2019-01-01. IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, SOUTH KOREA, Oct 27-Nov 02, 2019. p. 8738-8747. DOI : 10.1109/ICCV.2019.00883.
Backpropagation-Friendly Eigendecomposition2019-01-01. Conference on Neural Information Processing Systems (NeurIPS), Vancouver, CANADA, Dec 08-14, 2019.
Shape optimisation of technical devices via gradient descent using convolutional neural network proxiesUS2021157962 ; EP3679495 ; CN111295657 ; WO2019048085 . 2019.
Learning to Reconstruct Texture-less Deformable Surfaces from a Single View2018-03-23. International Conference on 3D Vision, Verona, Italy, September 5-8, 2018.
Method, system, and device for learned invariant feature transform for computer imagesUS10552709 ; US2018096224 . 2018.
Every Smile is Unique: Landmark-Guided Diverse Smile Generation2018-01-01. 31st IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, Jun 18-23, 2018. p. 7083-7092. DOI : 10.1109/CVPR.2018.00740.
LF-Net: Learning Local Features from Images2018. Neural Information Processing Systems (NIPS), Montréal Canada.
Geodesic Convolutional Shape Optimization2018.
Learning to Reconstruct Texture-less Deformable Surfaces from a Single View2018-01-01. 6th International Conference on 3D Vision (3DV), Verona, ITALY, Sep 05-08, 2018. p. 606-615. DOI : 10.1109/3DV.2018.00075.
The effects of aging on neuropil structure in mouse somatosensory cortex-A 3D electron microscopy analysis of layer 1PLOS ONE. 2018. DOI : 10.1371/journal.pone.0198131.
Robust 3D Object Tracking from Monocular Images Using Stable PartsTransactions on Pattern Analysis and Machine Intelligence (PAMI). 2018. DOI : 10.1109/TPAMI.2017.2708711.
A versatile calibration procedure for portable coded aperture gamma cameras and RGB-D sensorsNUCLEAR INSTRUMENTS AND METHODS IN PHYSICS RESEARCH SECTION A-ACCELERATORS SPECTROMETERS DETECTORS AND ASSOCIATED EQUIPMENT. 2018. DOI : 10.1016/j.nima.2017.12.065.
WILDTRACK: A Multi-camera HD Dataset for Dense Unscripted Pedestrian Detection2018. Proceedings of the IEEE international conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, Jun 18-23, 2018. p. 5030-5039. DOI : 10.1109/CVPR.2018.00528.
Geodesic Convolutional Shape Optimization2018. International Conference on Machine Learning (ICML), Stockholm, Sweden, July, 2018. p. 472-481.
Imposing Hard Constraints on Deep Networks: Promises and Limitations2017. CVPR Workshop on Negative Results in Computer Vision, Hawaii, HI, 2017.
Learning Active Learning from Data2017. Conference on Neural Information Processing Systems (NIPS).
Non-Markovian Globally Consistent Multi-Object Tracking2017. 2017 IEEE International Conference on Computer Vision (ICCV). p. 2563-2573. DOI : 10.1109/ICCV.2017.278.
Deep Occlusion Reasoning for Multi-Camera Multi-Target Detection2017. p. 271-279. DOI : 10.1109/ICCV.2017.38.
Method, System and Device for Direct Prediction of 3D Body Poses from Motion Compensated SequenceUS2017316578 . 2017.
Geometric Graph Matching Using Monte Carlo Tree SearchTransactions on Pattern Analysis and Machine Intelligence (PAMI). 2017. DOI : 10.1109/Tpami.2016.2636200.
Simultaneous Recognition and Pose Estimation of Instruments in Minimally Invasive Surgery2017. International Conference on Medical Image Computing and Computer-Assisted Intervention (MCCAI), Quebec, Canada.
Stereo-vision three-dimensional reconstruction of curvilinear structures imaged with a TEMUltramicroscopy. 2017. DOI : 10.1016/j.ultramic.2017.08.010.
Three-dimensional electron imaging and reconstruction of dislocations from a single acquisition2017. Microscopy Conference, Lausanne, Switzerland.
Learning Lightprobes for Mixed Reality Illumination2017. International Symposium on Mixed and Augmented Reality (ISMAR), Nantes, France, October 9–13, 2017.
Three-dimensional electron imaging of dislocations from a single sample tilt2016. The 16th European Microscopy Congress 2016, Lyon, France, 28 August - 2 September, 2016. DOI : 10.1002/9783527808465.
Do We Need Binary Features for 3D Reconstruction?2016. Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Las Vegas, NV, June, 2016. DOI : 10.1109/Cvprw.2016.144.
Analyzing Volleyball Match Data from the 2014 World Championships Using Machine Learning Techniques2016. Conference on Knowledge Discovery and Data Mining, San Francisco, CA, August, 2016.
Principled Parallel Mean-Field Inference for Discrete Random Fields2016. p. 5848-5857. DOI : 10.1109/CVPR.2016.630.
Learning to Assign Orientations to Feature Points2016. Computer Vision and Pattern Recognition (CVPR), Las Vegas, Nevada, USA, June 26-July 1, 2016.
Globally Optimal Cell Tracking using Integer Programming2016
Principled Parallel Mean-Field Inference for Discrete Random Fields2016. Computer Vision and Pattern Recognition (CVPR), Las Vegas.
Systems and methods for tracking interacting objectsUS9794525 ; US2015281655 . 2015.
Special Section: Pose & GestureComputer Vision And Image Understanding. 2015. DOI : 10.1016/j.cviu.2015.10.011.
Kullback-Leibler Proximal Variational Inference2015. Advances in Neural Information Processing Systems (NIPS), Montreal, Canada, December 9, 2015.
Kullback-Leibler Proximal Variational Inference2015.
A Novel Representation of Parts for Accurate 3D Object Detection and Tracking in Monocular Images2015. International Conference on Computer Vision (ICCV), Santiago, Chile, December 13-16, 2015.
TILDE: A Temporally Invariant Learned DEtector2015. Computer Vision and Pattern Recognition (CVPR), Boston, Massachusetts, USA.
Detecting and Tracking Cells using Network Flow Programming2015
Efficient scanning for em based target localizationWO2014001157 ; US8588509 . 2014.
Refining Mitochondria Segmentation in Electron Microscopy Imagery with Active Surfaces2014. European Conference on Computer Vision (ECCV) Workshop on Non-Rigid Shape Analysis and Deformable Image Alignment, Zurich, Switzerland, September 6-12, 2014. p. 367-379. DOI : 10.1007/978-3-319-16220-1_26.
Dendritic tree extraction from noisy maximum intensity projection images in C-elegansBiomedical Engineering Online. 2014. DOI : 10.1186/1475-925X-13-74.
On Rendering Synthetic Images for Training an Object Detector2014
Exploiting Enclosing Membranes and Contextual Cues for Mitochondria Segmentation2014. International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Boston, Massachusetts, USA, September 2014.
Simultaneous Segmentation and Anatomical Labeling of the Cerebral Vasculature2014. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Boston, Massachusetts, USA, Sep. 14-19, 2014.
Fast Part-Based Classification for Instrument Detection in Minimally Invasive Surgery2014. International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Boston, Massachusetts, USA, September 14-18, 2014. p. 692-699. DOI : 10.1007/978-3-319-10470-6_86.
Receptive Fields Selection for Binary Feature DescriptionIEEE Transactions on Image Processing. 2014. DOI : 10.1109/TIP.2014.2317981.
Dense Methods for Image Alignment with an Application to 3D Tracking2014
Separable Filter Learning with Tensor Decomposition2013
Method and apparatus for multiple object tracking with k-shortest pathsUS8615107 ; US2013177200 . 2013.
Learning Separable Filters with Shared Parts2013
Tracking Multiple Handball Players using Multi-Commodity Network Flow for Assessing Tactical Behavior2013. Scientific Conference Women and Handball: Scientific and Practical Approaches, Vienna, Austria, November, 2013.
Non-Linear Domain Adaptation with Boosting2013. Neural Information Processing Systems (NIPS), Lake Tahoe, Nevada, USA, December 5-8, 2013.
Facial Descriptors for Identity-Preserving Multiple People Tracking2013
Flash Scanning Electron Microscopy2013. 16th Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Nagoya, Japan, September 22-26, 2013.
Tracklet-based Multi-Commodity Network Flow for Tracking Multiple PeopleEP2780871 ; EP2780871 ; WO2013072401 ; WO2013072401 . 2013.
Detection of Aircrafts on a Collision Course using Spatio-Temporal HOG2013
An Optimal Policy for Target Localization with Application to Electron Microscopy2013. International Conference on Machine Learning (ICML), Atlanta, GA, USA, June 16-21, 2013. p. 1-9.
Multi-camera face detection and recognition applied to people tracking2013
Tubular Geodesics using Oriented Flux: An ITK ImplementationInsight Journal. 2013.
KernelBoost: Supervised Learning of Image Features For Classification2013
Semi-Automated Reconstruction of Curvilinear Structures in Noisy 2D images and 3D image stacks2013
Tracking Multiple Players using a Single CameraMachine Vision and Applications. 2013.
Learning Image Descriptors with the Boosting-Trick2012. NIPS, Lake Tahoe, CA, USA, 2012.
Real time multi-object tracking using multiple cameras2012
Hybrid Algorithms for the Minimum-Weight Rooted Arborescence Problem2012. International Conference on Swarm Intelligence, Brussels, Belgium, September, 2012.
Transfer Learning by Sharing Support Vectors2012
On the Relevance of Sparsity for Image Classification2012
Structured Image Segmentation using Kernelized Features2012. European Conference on Computer Vision, Florence, Italy, October 2012.
Learning Separable Filters2012
Laplacian Meshes for Monocular 3D Shape Recovery2012. European Conference on Computer Vision, Florence, October 2012.
AUTOMATIC MAPPING FROM ULTRA-LIGHT UAV IMAGERY2012. EuroCOW 2012, Barceloa, Spain, February 8-10, 2012.
Turning Augmented Reality into a media: Design exploration to build a dedicated visual language2011-10-26. 2011 IEEE International Symposium on Mixed and Augmented Reality - Arts, Media, and Humanities, Basel, Switzerland, 26-29 October, 2011. p. 83-89. DOI : 10.1109/ISMAR-AMH.2011.6093661.
Automated Quantification of Morphodynamics for High-Throughput Live Cell Imaging Datasets2011. 1st International SystemsX.ch Conference on Systems Biology.
Morphodynamic profiling to explore spatio-temporal signaling networks regulating neurite outgrowth2011. 1st International SystemsX.ch Conference on Systems Biology.
myCopter – Enabling Technologies for Personal Aerial Transportation Systems2011. International HELI World Conference, Frankfurt/Main, Germany, 2011.
myCopter – Enabling Technologies for Personal Aerial Transportation Systems.2011. European Rotorcraft Forum, Vergiate/Gallarate, Italy, September 13-15, 2011.
Simplified Building Models Extraction From Ultra-Light UAV Imagery2011. UAV-g 2011 - Unmanned Aerial Vehicle in Geomatics, Zürich, CH, September 14-16, 2011.
The Accuracy of Automatic Photogrammetric Techniques on Ultra-light UAV Imagery2011. UAV-g 2011 - Unmanned Aerial Vehicle in Geomatics, Zürich, CH, September 14-16, 2011.
|R. Achanta, A. Shaj, K. Smith, A. Lucchi, P. Fua, and S. S�sstrunk.
IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 34, num. 11, p. 2274 - 2282, 2012.
|SLIC Superpixels Compared to State-of-the-art Superpixel Methods|
|F Fleuret, J Berclaz, R Lengagne, and P Fua
Pattern Analysis and Machine Intelligence, 30 (2), 267-282, 2008
|Multicamera People Tracking with a Probabilistic Occupancy Map|
|V. Lepetit and P. Fua
Transactions on Pattern Analysis and Machine Intelligence, Vol. 28, Nr. 9, pp. 1465--1479, 2006.
|Keypoint Recognition using Randomized Trees|
|P. Fua and Y. G. Leclerc
International Journal of Computer Vision, Vol. 16, pp. 35-56, 1995.
|Object-Centered Surface Reconstruction: Combining Multi-Image Stereo and Shading|
Machine Vision and Applications, Vol. 6, Nr. 1, pp. 35-49, 1993.
|A Parallel Stereo Algorithm that Produces Dense Depth Maps and Preserves Image Features|
Teaching & PhD
Doctoral program in computer and communication sciences
Doctoral Program in Electrical Engineering