Mathieu Salzmann
Adjunct Professor
Works on deep learning for 2D and 3D scene understanding, efficient and robust deep learning, and domain adaptation.
Mathieu Salzmann's lab at EPFL works on 3D vision & neural rendering, recognition, detection & segmentation and signal processing & information theory. 35 papers since 2023, 20 at top venues, mostly CVPR and ICCV. 10 PhD students. 12 alumni. Applications close .
At a glance
- papers, 2023–2026
- 9/ year
- at T1 venues
- 57%
- PhD students
- 10
- MS students
- 0
- alumni listed
- 12
Papers by yearfrom the lab homepage, 2023–2026
- 3 papers2026 · T1 0 · collecting
- 6 papers2025 · T1 2
- 11 papers2024 · T1 6
- 15 papers2023 · T1 12
T1 venuesOtherVenues: CVPR, ICCV, TPAMI, WACV, 3DV
Research focuspaper titles that use each topic's words
- Recognition, detection & segmentation6 papers
- 3D vision & neural rendering4 papers
- Signal processing & information theory2 papers
- robust deep0 papers
- domain adaptation0 papers
Paper list
From the lab homepage.
20263 papers
- WACVFastPose-ViT: A Vision Transformer for Real-Time Spacecraft Pose Estimation (opens in a new tab)
- International Sports Analytics Conference and ExhibitionShot and Technical Error Detection in Tennis Using Graph Convolutional Networks (opens in a new tab)
- IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSSubtractive Modulative Network with Learnable Periodic Activations (opens in a new tab)
20256 papers
- IEEE Intelligent Vehicles Symposium6Img-to-3D: Few-Image Large-Scale Outdoor Novel View Synthesis (opens in a new tab)
- International Conference on Computer Vision and Image ProcessingJoint Compression of Multi-model Systems for Edge Devices (opens in a new tab)
- ECML-PKDDLeveraging Gradient Information for Out-of-Domain Performance Estimations (opens in a new tab)
- ICCV (T1)Self-Ensembling Gaussian Splatting for Few-Shot Novel View Synthesis (opens in a new tab)
- ECCV (T1)SINDER: Repairing the Singular Defects of DINOv2 (opens in a new tab)
- Acta AstronauticaTiny satellites, big challenges: A feasibility study of machine vision pose estimation for PocketQubes during conjunctions (opens in a new tab)
202411 papers
- TPAMI (T1)A Closed-Form Pairwise Solution to Local Non-Rigid Structure-from-Motion (opens in a new tab)
- IEEE/CVF Conference on Computer Vision and Pattern Recognition CVPRCLOAF: CoLlisiOn-Aware Human Flow (opens in a new tab)
- TPAMI (T1)Detecting Road Obstacles by Erasing Them (opens in a new tab)
- IEEE/CVF Conference on Computer Vision and Pattern Recognition CVPRDvmnet: Computing Relative Pose for Unseen Objects Beyond Hypotheses (opens in a new tab)
- NeurIPS (T1)Generalize or Detect? Towards Robust Semantic Segmentation Under Multiple Distribution Shifts (opens in a new tab)
- CVPR (T1)HOISDF: Constraining 3D Hand-Object Pose Estimation with Global Signed Distance Fields (opens in a new tab)
- —HOISDF: Constraining 3D Hand-Object Pose Estimation with Global Signed Distance Fields: Processed data and trained models (opens in a new tab)
- TPAMI (T1)Match Normalization: Learning-Based Point Cloud Registration for 6D Object Pose Estimation in the Real World (opens in a new tab)
- Journal of Data Mining & Digital HumanitiesToward Automatic Typography Analysis: Serif Classification and Font Similarities (opens in a new tab)
- 3DVUnsupervised 3D Keypoint Discovery with Multi-View Geometry (opens in a new tab)
- TMLR (T1)Using Motion Cues to Supervise Single-Frame Body Pose and Shape Estimation in Low Data Regimes (opens in a new tab)
202315 papers
- ICCV (T1)AutoSynth: Learning to Generate 3D Training Data for Object Point Cloud Registration (opens in a new tab)
- WACVCenter-aware Adversarial Augmentation for Single Domain Generalization (opens in a new tab)
- ICCV (T1)Center-Based Decoupled Point Cloud Registration for 6D Object Pose Estimation (opens in a new tab)
- CVPR (T1)CLIP the Gap: A Single Domain Generalization Approach for Object Detection (opens in a new tab)
- IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) WorkshopDense Multitask Learning to Reconfigure Comics (opens in a new tab)
- CVPR (T1)DrapeNet: Garment Generation and Self-Supervised Draping (opens in a new tab)
- CVPR (T1)Knowledge Distillation for 6D Pose Estimation by Aligning Distributions of Local Predictions (opens in a new tab)
- CVPR (T1)Learning Transformations To Reduce the Geometric Shift in Object Detection (opens in a new tab)
- ICCV (T1)Linear-Covariance Loss for End-to-End Learning of 6D Pose Estimation (opens in a new tab)
- RA-LPerspective Aware Road Obstacle Detection (opens in a new tab)
- CVPR (T1)Rigidity-Aware Detection for 6D Object Pose Estimation (opens in a new tab)
- CVPR (T1)Spatiotemporal Self-supervised Learning for Point Clouds in the Wild (opens in a new tab)
Where alumni went12 listed
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
- CNPeking
Chao Xu
- Video understanding
- USCMU
Alexander Hauptmann
- Video understanding
- USUMD
Christopher Metzler
- Computational imaging & photography
- USUIUC
Derek Hoiem
- Recognition, detection & segmentation
- USUT Austin
Chen Yu
- Video understanding
- CNTsinghua
Chun Yuan
- Recognition, detection & segmentation
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- This site is shared with other labs, so counts may overlap.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
This lab shares a website with other labs. Member and alumni counts may include them.
Tiers are venue classifications, not lab rankings.
EPFLcloses