OphNet2024: Ophthalmic Surgical Video Dataset

Large-scale multi-procedure ophthalmic surgical video dataset covering 66 surgery types, 102 phases, 150 operations (~285 h). 1,969 untrimmed videos; 17,508 trimmed operation-level clips; 14,674 trimmed phase-level clips across cataract, vitreoretinal, glaucoma, corneal, refractive, oculoplastic, and strabismus. 743 videos have time-boundary annotations. ECCV 2024.

At a glance

Field	Value
Short name	ophnet2024
Full name	OphNet2024: Ophthalmic Surgical Video Dataset
First published	2024-10
Publication date precision	month
Publication date evidence	raw.githubusercontent.com/minghu0830
Publication date source field	Official project README News: Oct 2024 dataset release
Publication date reviewed	2026-09-11
Primary category	surgical_video
Resource role	current_dataset
Dataset family	ophnet2024
Contained modalities	surgical_video
Tasks	classification, phase_recognition, detection
Primary reported quantity	2,278 videos
Classes	8 (Cataract Surgery, Vitreoretinal Surgery, Glaucoma Surgery, Corneal Surgery, Refractive Surgery, Oculoplastic Surgery, Strabismus Surgery, Other)
Splits	train, val, test
Size	583.0 GB
Source-stated terms	CC BY-NC-SA 4.0
Normalized terms	cc-by-nc-sa
Descriptive screening label	Explicit noncommercial clause recorded; check source
Terms scope	dataset_files
Access friction	self_service_authenticated
Route backend	HuggingFace Hub
Availability	available (checked 2026-07-21)
Acquisition support	standard_platform_supported
Legacy sample-loader status	Standard loader included

Reported quantities

Role	Count	Unit	Scope	Basis	Evidence
Primary	2,278	videos	Primary quantity reported in the reviewed catalog source	legacy_catalog_field	huggingface.co/datasets
Additional	1,969	videos	Untrimmed videos	official_source_description	huggingface.co/datasets
Additional	17,508	video_clips	Trimmed operation-level clips	official_source_description	huggingface.co/datasets
Additional	14,674	video_clips	Trimmed phase-level clips	official_source_description	huggingface.co/datasets

Counts retain their source-reported units. Additional rows can describe components, paired items, or derivative copies and are not automatically added to the primary quantity.

Notes

~583 GB total: untrimmed ~305 GB, trimmed operations ~139 GB, trimmed phases ~139 GB, features ~26 GB. num_samples = 2,278 source videos. May require HF_TOKEN for gated access. Set HF_TOKEN in .env. GitHub: https://github.com/minghu0830/OphNet-benchmark

Access information and download

CLI

# Read-only preflight
eyehub download ophnet2024 --data-dir ./data --dry-run --json

# Download, only when preflight reports supported behavior
eyehub download ophnet2024 --data-dir ./data

Python

from eyedatahub.acquisition import preflight_dataset
from eyedatahub.datasets.registry import REGISTRY

ds = REGISTRY.get_dataset('ophnet2024')
print(preflight_dataset(ds, './data'))  # no download

Upstream page: huggingface.co/datasets

Source-term evidence: huggingface.co/datasets

Loader example

This entry includes a standard DatasetSample loader.

from pathlib import Path
from eyedatahub.datasets.registry import REGISTRY

data_dir = Path('~/.eyedatahub/data').expanduser()
ds = REGISTRY.get_dataset('ophnet2024')
samples = ds.load(data_dir, split='test')
for s in samples[:5]:
    print(s.sample_id, s.label, s.image_path)

Citation

BibTeX

@misc{ophnet2024,
  title  = { OphNet2024: Ophthalmic Surgical Video Dataset },
  note   = { Hu M et al., 'OphNet: A Large-Scale Video Benchmark for Ophthalmic Surgical Workflow Understanding', ECCV 2024. arXiv:2406.07471. HuggingFace: https://huggingface.co/datasets/xioamiyh/OphNet2024 — GitHub: https://github.com/minghu0830/OphNet-benchmark },
  year   = { 2024 },
  url    = { https://huggingface.co/datasets/xioamiyh/OphNet2024 },
}

Plain text

Hu M et al., 'OphNet: A Large-Scale Video Benchmark for Ophthalmic Surgical Workflow Understanding', ECCV 2024. arXiv:2406.07471. HuggingFace: https://huggingface.co/datasets/xioamiyh/OphNet2024 — GitHub: https://github.com/minghu0830/OphNet-benchmark

Source-stated terms

• Raw source string: CC BY-NC-SA 4.0
• Normalized category: cc-by-nc-sa
• Apparent scope: dataset_files
• Descriptive screening label: Explicit noncommercial clause recorded; check source

⚠️ Source-stated terms, scope, and normalized labels are curation metadata, not legal advice or a permission finding. Review the current official source before transfer or reuse.

Similar resources by shared modality

• ophora: Ophora-160K: Ophthalmic Surgical Video Instruction Dataset (162,185 video clip instruction pairs, unknown)
• lmod_plus: LMOD+ Multimodal Ophthalmology Benchmark (32,633 annotated instances, unknown)
• insegcat: InSegCat: Instance Segmentation for Cataract Surgery (5,581 annotated images, research-only)
• cadis: CaDIS: Cataract Dataset for Image Segmentation (4,670 images, research-only)
• cataract_lmm: Cataract-LMM: Large-Scale Multi-Source Multi-Task Cataract Surgery Benchmark (3,000 videos, cc-by-nc-nd)
• lmod_cataract_1k: LMOD-Cataract-1K (2,256 images, cc-by)
• lmod_cataract_1k_cot: Cataract-1K Surgical Analysis Chain-of-Thought Dataset (2,256 images, mit)
• cataract1k: Cataract-1K: Large-Scale Cataract Surgery Video Dataset (1,000 videos, research-only)