AI4M3D 2026

Deadlines
Computer Vision/CORE Unranked

AI4M3D 2026

Artificial Intelligence for Medical 3D Vision Workshop

1788847200000Malmö, SwedenOfficial workshop site Site reachable

AI4M3D 2026 is a workshop held in conjunction with ECCV 2026 that focuses on artificial intelligence for medical 3D vision, bringing together researchers working at the intersection of AI, computer vision, and medical imaging. The workshop aims to advance techniques in 3D reconstruction, generative AI, neural rendering, and XR-based visualization for clinical applications, while addressing challenges like data scarcity, anatomical variability, and clinical reliability.

Official CFP Back to deadlines Verified September 9, 2026

Paper fit

Contribution paths

A strong submission should clearly identify its contribution and evaluate it appropriately.

Full Papers

Original, previously unpublished work submitted for peer review; must follow ECCV 2026 submission guidelines, limited to 14 pages plus references; accepted papers will be published in ECCV 2026 proceedings.

Previously Submitted Papers

Papers previously submitted to other venues are welcome, but they will not be included in the ECCV 2026 proceedings.

Research areas in scope

01

AI for Medical 3D Reconstruction and Neural Rendering

AI for medical 3D reconstruction and neural rendering3D reconstruction from CT or MRI data3D Gaussian Splatting for medical scene understandingPoint cloud and mesh reconstruction for surgical planningReal-time intraoperative 3D rendering4D dynamic reconstruction from medical images
02

Generative AI for Medical and Neural 3D Data

Generative AI for medical and neural 3D data3D models generation from EEG or MRI data3D shape generation and completion for anatomyGenerative AI for 3D medical data synthesisNeural data-driven 3D brain representation generationInteractive 3D avatars for medical training
03

Vision-Language Models for 3D Understanding

Vision-language models for 3D understandingLanguage-promptable 3D segmentation for anatomical structuresEfficient 3D VLM architectures for volumetric dataMulti-modal 3D clinical reasoning and visual question answeringZero-shot and few-shot 3D understanding via foundation VLMs
04

XR-Supported Visualization and Simulation

XR-supported visualization and simulationVirtual overlays for intraoperative guidanceXR interfaces for medical education and surgical trainingCollaborative multi-user XR for remote clinical consultationAI-based rendering and scene understanding for XR
05

Foundation and Self-Supervised Models for Medical 3D Vision

Foundation and self-supervised models for medical 3D visionSelf-supervised pretraining on 3D medical volumesUniversal anatomical segmentation with foundation modelsParameter-efficient fine-tuning for 3D clinical tasks with minimal labeled dataContinual and federated pretraining of 3D foundation modelsLearning from sparse, noisy, or heterogeneous medical dataSemi-supervised and few-shot 3D segmentation with foundation model priorsFederated learning for multi-view 3D imaging
06

Uncertainty Quantification and Label-Efficient Learning

Uncertainty quantification for clinical deployment of 3D modelsLabel-efficient learning from noisy and partially annotated 3D medical data

Policies worth checking twice

  • All papers undergo double-blind peer review with at least two reviewers.
  • Submissions must be in PDF format following the ECCV 2026 Submission Template.
  • Manuscripts are limited to 14 pages; additional pages for references are allowed.
  • Supplemental materials in PDF format are permitted.
  • Accepted papers will be published as part of the ECCV 2026 proceedings.
  • Authors must comply with ECCV 2026's ethics guidelines, recommended best practices, and FAQs.
  • Submissions must be original and previously unpublished to be included in the proceedings.

Official sources

Compiled from the official call for papers. The organizers’ pages remain authoritative.

Last verified September 9, 2026