Skip to content

ECCV 2026 — Program Notes (pre-meeting prep)

Last updated: May 11, 2026 — T-17 weeks. Submission cycle is mid-flight: paper submissions closed Mar 5; rebuttal closed today (May 11); final decisions land Jun 17, 2026; camera-ready Jun 26. Workshop proposals were accepted Apr 12, with individual workshop sites coming online through May–August. Accepted-paper list not yet public — meaningful keyword scanning of accepts cannot begin until late June.

This page exists to lock the scope filter, workshop slate, anticipated paper themes, and cross-vault pairings before any tool extraction begins. It is intentionally explicit about what is out of scope — ECCV's bulk is general computer vision, and conflating that bulk with the medical-imaging / pathology-FM subset would destroy the vault's signal-to-noise.

Verified meeting facts

  • Dates: Tue September 8 – Sat September 12, 2026.
  • Sep 8 (Tue): Workshops + Tutorials Day 1 — anticipated medical / bio cluster here.
  • Sep 9 (Wed): Workshops + Tutorials Day 2 — second day of workshop / bio cluster.
  • Sep 10 (Thu) – Sep 12 (Sat): Main conference, orals + poster sessions + Expo.
  • Venue: Malmö Arena and Malmömässan, Malmö, Sweden. (Malmö is the southern Swedish city across the Öresund Bridge from Copenhagen; easy travel pairing with Copenhagen Airport CPH.)
  • Scale: Anticipated ~6,700–7,500 attendees, building on ECCV 2024 Milan's 6,700+ (smaller than CVPR's ~10,000 but the same overall venue tier).
  • Proceedings: Springer LNCS (paywalled at the DOI, but mirrored open-access via ECVA at ecva.net/papers/eccv_2026/) — typically posts within ~1–2 weeks of meeting. OpenReview venue: thecvf.com/ECCV/2026/Conference.
  • Biennial cadence: ECCV runs even years; alternates with ICCV (odd years). 36th edition (the "European" cycle dating back to 1990 with two-year gaps).

Scope filter (load-bearing — repeat from tools/index.md)

ECCV is scope-limited by design in this vault. The filter is:

  1. Medical / biomedical workshop (slate still being announced through May–August; anticipated 2–4 medical / bio workshops based on ECCV 2022 + 2024 history).
  2. Main-track keyword match on medical / pathology / radiology / oncology / clinical / biomedical / bio-imaging / segmentation-for-medicine vocabulary.
  3. Architectural building block with a documented or near-certain medical adoption path — DINOv3-class SSL pretraining, SAM-3-class segmentation FM, next-gen CLIP / SigLIP multimodal alignment, slide-level aggregation, diffusion / flow-matching for medical synthesis.

Everything else stays out. The light-build pattern mirrors CVPR 2026, ICML 2026, and NeurIPS 2026.

Medical / biomedical workshop slate (partial — being announced)

Workshop proposals were accepted Apr 12, 2026; individual workshop sites are coming online through May–August. The slate below is partial and based on anticipation from ECCV 2018 / 2020 / 2022 / 2024 history + the CVPR 2026 medical workshop pattern; expect revision as the official slate populates.

Workshop Acronym Likely date Format Notes
BioImage Computing BIC Sep 8 or 9 (TBD) Half / full day Recurring at ECCV 2018 / 2020 / 2022 / 2024; bio-imaging methods, microscopy, cell segmentation, deep learning for bio-imaging — anticipated for 2026
Medical Computer Vision (ECCV edition) MCV Sep 8 or 9 (TBD) Half / full day ECCV-edition ran at 2022; cross-pollinates with CVPR MCV — anticipated
Foundation Models for Medical Vision (ECCV variant) FMV-ECCV Sep 8 or 9 (TBD) Half day Speculative — possible ECCV companion to the CVPR FMV series
Multimodal Foundation Models for Biomedicine (ECCV variant) MMFM-BIOMED-ECCV Sep 8 or 9 (TBD) Half day Speculative — possible ECCV companion to CVPR MMFM-BIOMED
CV4Good — Computer Vision for Humanitarian Action CV4Good Sep 8–9 TBD Confirmed; Good AI Lab + MSF; humanitarian / global-health imaging — partial vault relevance

Workshop slate revision will land when eccv.ecva.net/Conferences/2026/Workshops is fully populated (anticipated June–July, in line with prior ECCV cycles).

Anticipated paper themes (5 high-signal axes)

These are the patterns we expect to see repeatedly in both main-track and workshop accepts. We will not commit to a full theme synthesis page (that would be out of scope for a light build), but the per-tool entries will tag against these themes:

  1. Pathology foundation model successors and slide-level aggregators. Next-generation tile encoders (DINOv3-class, scaling beyond UNI / GigaPath / Virchow), slide-level aggregation transformers (ABMIL / TransMIL successors), multi-cohort pretraining at TCGA / CPTAC / European-hospital-network scale (Charité, Karolinska, Erasmus MC, Institut Curie, IEO Milan, NKI). Direct upstream of CHIEF / Prov-GigaPath / UNI updates. European pathology consortia (BIGPICTURE, EOSC4Cancer, EuCAIM) anchor papers here that don't appear at CVPR.

  2. Pathology and radiology vision-language models. CONCH / PathChat / MUSK / Virchow successors on the pathology side; MAIRA / RadFM / Merlin / CXR-LLaVA successors on the radiology side. Pathology-report alignment, radiology-report generation, prompt-tunable report drafting. Knowledge-enhanced visual-language pretraining (KEP-line, ECCV 2024) is the canonical anchor. Expect the speculative MMFM-BIOMED-ECCV (if it materializes) to be the workshop home for the bulk of these.

  3. Generative medical imaging and counterfactuals. Latent diffusion / DiT / flow-matching priors for MRI / CT reconstruction; ControlNet-style anatomy-conditioned synthesis; counterfactual generation for fairness audits and explainability; synthetic-data augmentation for rare-cohort segmentation. Bridges to MICCAI 2026 DGM4MICCAI workshop (2–3 weeks later). European diffusion-research community (LMU Compvis, EPFL, IST Austria) anchors many of these papers.

  4. Segmentation foundation models — SAM line and successors, plus bio-imaging segmentation. SAM-3 / EfficientSAM-2 / MedSAM-class adapters; promptable segmentation for tumor / lesion / organ tasks; few-shot and zero-shot transfer to BraTS / HECKTOR / KiTS / AutoPET / PANORAMA. Plus the BioImage Computing axis: 2D/3D cell segmentation, subcellular structure segmentation, microscopy super-resolution, instance segmentation for high-content screens — directly relevant to cancer biology pipelines (IHC quantification, tumor microenvironment imaging).

  5. Multimodal medical reasoning, federated learning, and EU regulatory-grade clinical VLMs. Chain-of-thought reasoning over chest X-rays / pathology slides, agentic clinical workflows, retrieval-augmented diagnosis, multimodal alignment between text reports + imaging + structured EHR. The ECCV-distinctive thread: federated learning across European hospital networks (PHAROS, EuCAIM, OPTIMAL, EOSC4Cancer) and EU AI Act / MDR-compliant deployment frameworks — work that doesn't surface at CVPR. Cross-link to AACR 2026 agentic-AI axis and ICML 2026 agentic-AI subset.

Out-of-scope themes (explicitly skipped)

To keep the filter honest, the following ECCV 2026 themes get no coverage in this vault even when they are technically impressive:

  • General scene understanding, depth estimation, optical flow, monocular 3D — no medical thread.
  • Autonomous driving perception (KITTI / Waymo / nuScenes-class) — no medical thread.
  • Robotics manipulation, embodied AI, simulation-to-real — no medical thread (surgical robotics is MICCAI's CAI track, not ECCV's general robotics track).
  • General video generation (Sora / Veo / Movie Gen class) — no medical thread unless medical-imaging-specific.
  • Face recognition, person re-identification, biometrics — no medical thread.
  • AR / VR / NeRF / Gaussian splatting (general) — no medical thread.
  • Adversarial robustness and watermarking for general images — out of scope.
  • General efficiency / quantization / pruning — out of scope unless the paper benchmarks on a medical FM.
  • General-purpose CV4Good humanitarian-imaging entries that aren't health / medical — partial CV4Good relevance only.

Cross-vault pairings (the load-bearing chains)

The point of carrying ECCV in this vault is the cross-conference arcs. The ECCV–MICCAI overlap is the tightest of any conference pair in the vault — 2–3 weeks separates them in 2026, both in Europe (Malmö → Daejeon for MICCAI, but European lab authorship dominates both). Four canonical chains to track:

  1. ECCV pretraining recipe → MICCAI methods (2–3 weeks later) → AACR translational poster. A new SSL recipe at ECCV Sep 8–12 → pathology-FM v2 paper at MICCAI 2026 Sep 27–Oct 1 → translational poster at AACR 2027 April. The ECCV entry should explicitly link forward to the MICCAI cousin and the eventual AACR poster.

  2. CVPR (Jun) → ECCV (Sep) → MICCAI (Sep–Oct) European mirror arc. A US-led CVPR pretraining recipe in June 2026 → European-lab response / variant at ECCV three months later → MICCAI methods application within 2–3 weeks. Tracking the CVPR–ECCV–MICCAI triple is informative for understanding the US / European division of method labor in medical-imaging FM development.

  3. ECCV medical-VLM → RSNA clinical-deployment readout. A new radiology VLM at ECCV → vendor demonstration or prospective-study readout at RSNA 2026 Nov 29–Dec 3. The 11-week arc is reliable for diagnostic-imaging VLMs.

  4. ECCV SAM successor → MICCAI challenge winner. A SAM-3-class segmentation FM at ECCV → adoption in BraTS / HECKTOR / KiTS / AutoPET / PANORAMA winning entry at MICCAI 2026 within the same month. Track which ECCV segmentation paper anchors each MICCAI challenge result.

What is NOT being built

To be explicit (light-build hygiene):

  • No digest/ directory. No per-day digests. The general-vision bulk dominates per-day proceedings and would dilute signal.
  • No themes.md. The five themes above are scoped here in program-notes.md and tagged per tool entry; a separate synthesis page is not warranted.
  • No keynotes/. ECCV plenaries are general-vision; we cover them only if a known medical-imaging / pathology-FM author is on stage.
  • No mkdocs.yml. This vault piggybacks on the parent site config (per repo convention for sibling conference vaults).
  • No timeline.md. Conference timeline lives at ../../timeline.md; no per-vault duplication.

Next actions

  • May 11 – Jun 17, 2026 (now → final decisions, T-17 → T-12 weeks): workshop sites continue posting; track accepted workshops as they post (especially confirm/deny BIC, MCV-ECCV, FMV-ECCV, MMFM-BIOMED-ECCV anticipation). No main-track filtering possible until Jun 17.
  • Jun 17 – Jul 31, 2026 (T-12 → T-6 weeks): accept list goes public Jun 17. Begin scanning with keyword filter; pre-stub the ~20–40 high-signal entries. Camera-ready Jun 26; arXiv mirrors land through July. Cross-check authorship against the MICCAI 2026 accept list (already posted May 7) for cousin-paper pairing — this is the single most important cross-link operation for this vault given the 2–3 week ECCV–MICCAI gap.
  • Aug 2026 (T-5 → T-2 weeks): workshop accept lists post on individual workshop sites. Pull DOIs and arXiv IDs into stub pages.
  • Sep 8–12, 2026 (during the meeting): opportunistic capture of slide deposits, GitHub repos, demo links, social reactions. Sep 8–9 workshop cluster is the densest pair of days for vault filter. Plan for dual-coverage workflow — MICCAI 2026 (Sep 27–Oct 1) starts 15 days after ECCV ends; the same labs will be at both meetings.
  • Sep 13 – Sep 30, 2026 (post-meeting): ECVA Open Access proceedings post ~1–2 weeks after meeting. Bulk-populate tools/ with the filtered subset, consolidating ECCV–MICCAI cousin pairs as MICCAI runs concurrently. Cross-link into AACR pathology-FM dossiers.

Sources