Full Paper
Main text must be 4–8 pages (including figures and tables), with unlimited reference pages and no page limit for technical appendices. Maximum file size is 50MB. Supplementary material up to 100MB allowed.
Computer Vision/CORE Unranked
CVPR 2026
CVPR 2026 Workshop on 3D Geometry Generation for Scientific Computing
The 3D4Science workshop at CVPR 2026 focuses on advancing 3D and 4D geometry generation for scientific computing, bringing together computer vision researchers and domain experts to address challenges in reconstructing complex, multi-scale scientific data. The workshop aims to foster interdisciplinary collaboration on scalable, accurate, and efficient methods for applications in climate science, biomedicine, urban planning, and other scientific domains.
Key deadlines
Verified September 9, 2026Abstract registration
April 6, 2026
AoE
Full paper
April 8, 2026
AoE
Workshop timeline
Submission and decisions
Abstract registrationKey deadline
April 6, 2026 · AoE
Full paperKey deadline
April 8, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Research areas in scope
01
Technical Challenges in 3D/4D Reconstruction
Multi-scale PatternsLarge-scale ScenesHeterogeneous ViewsDynamic and Time-varying ViewsComplex and Unstructured GeometriesOcclusions and Missing ObservationsComputational ComplexityGeneralization and Scene AdaptabilityReal-time Rendering for Dynamic ScenesLighting and Viewpoint Variations
Policies worth checking twice
- Submissions must be anonymized for double-blind review; no identifying information in main text, supplementary material, or linked code.
- Authors must not use 'our previous work'—instead, anonymize citations (e.g., 'Smith et al. [1]').
- No acknowledgements allowed in submitted papers.
- Abstract submission is mandatory by April 5, 2026; author list cannot be changed after this date without approval.
- Author list may be reordered after abstract deadline; additions or removals require written justification and approval by program chairs.
- Supplementary material (code, data) must be anonymized and is reviewed at the discretion of reviewers.
- Use of LLMs must be disclosed if integral to methodology; authors are fully responsible for all content.
- Submissions must be original, unpublished work, and in English.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026