SurgAtlas is a new dataset of 15,291 surgical videos totaling 2,391 hours with multi-level annotations that supports finetuning models to competitive performance on surgical benchmarks.
arXiv preprint arXiv:2001.11190 (2020)
10 Pith papers cite this work, alongside 119 external citations. Polarity classification is still indexing.
10
Pith papers citing it
119
external citations · external index
citation-role summary
dataset 1
method 1
citation-polarity summary
representative citing papers
A hierarchical prompt tree with self-reflection graph propagation enables positive forward and backward knowledge transfer in incremental surgical instrument segmentation, improving over baselines by more than 5% and 11% on two benchmarks.
Introduces the first publicly accessible native 4K resolution endoscopic video dataset for robotic-assisted minimally invasive procedures.
FAROS uses flow-guided propagation from zero-shot masks and optical flow to create dense temporally consistent labels from sparse keyframes, improving joint multi-task learning across temporal and spatial surgical tasks on GraSP, MISAW, and AutoLaparo.
RoboSurg-VQA is a new segmentation-aware VQA benchmark created by repurposing public surgical datasets with fixed clinically motivated questions and closed answer sets.
SAM 2 delivers more accurate video segmentation with 3x fewer user interactions and 6x faster image segmentation than the original SAM by training a streaming-memory transformer on the largest video segmentation dataset collected to date.
A multi-frame network with SAM 3-derived mask priors achieves 72.4% F1 tip and 58.0% F1 anchor localization in surgical videos without manual mask annotations for training.
A new synthetic dataset and geometry-consistent dense correspondence framework improve RGB-only pose estimation accuracy for surgical instruments on three evaluation datasets.
Releases the SurgVU dataset of surgical videos and labels to enable machine learning research in surgical data science.
SegSTRONG-C provides a new benchmark where top models reach 0.9394 DSC and 0.9301 NSD on corrupted surgical tool segmentation tests, showing conventional techniques help but calling for more innovative robustness methods.
citing papers explorer
-
SurgAtlas: A Large-Scale Surgical Video-Language Dataset with 2,391 Hours of Open and Minimally Invasive Surgery
SurgAtlas is a new dataset of 15,291 surgical videos totaling 2,391 hours with multi-level annotations that supports finetuning models to competitive performance on surgical benchmarks.
-
Unlocking Positive Transfer in Incrementally Learning Surgical Instruments: A Self-reflection Hierarchical Prompt Framework
A hierarchical prompt tree with self-reflection graph propagation enables positive forward and backward knowledge transfer in incremental surgical instrument segmentation, improving over baselines by more than 5% and 11% on two benchmarks.
-
SurgiSR4K: A High-Resolution Endoscopic Video Dataset for Robotic-Assisted Minimally Invasive Procedures
Introduces the first publicly accessible native 4K resolution endoscopic video dataset for robotic-assisted minimally invasive procedures.
-
Temporally Consistent Label Interpolation for Robust Surgical Multi-Task Learning under Challenging Conditions
FAROS uses flow-guided propagation from zero-shot masks and optical flow to create dense temporally consistent labels from sparse keyframes, improving joint multi-task learning across temporal and spatial surgical tasks on GraSP, MISAW, and AutoLaparo.
-
RoboSurg-VQA: A Multimodal Benchmark for Surgical Segmentation-Aware Visual Question Answering
RoboSurg-VQA is a new segmentation-aware VQA benchmark created by repurposing public surgical datasets with fixed clinically motivated questions and closed answer sets.
-
SAM 2: Segment Anything in Images and Videos
SAM 2 delivers more accurate video segmentation with 3x fewer user interactions and 6x faster image segmentation than the original SAM by training a streaming-memory transformer on the largest video segmentation dataset collected to date.
-
Dense Structural Priors for Sparse Functional Landmark Localization in Surgical Videos
A multi-frame network with SAM 3-derived mask priors achieves 72.4% F1 tip and 58.0% F1 anchor localization in surgical videos without manual mask annotations for training.
-
SurfSurg6D: Geometry Consistent Dense Correspondence for Textureless Surgical Instrument Pose Estimation
A new synthetic dataset and geometry-consistent dense correspondence framework improve RGB-only pose estimation accuracy for surgical instruments on three evaluation datasets.
-
Surgical Visual Understanding (SurgVU) Dataset
Releases the SurgVU dataset of surgical videos and labels to enable machine learning research in surgical data science.
-
SegSTRONG-C: Segmenting Surgical Tools Robustly On Non-adversarial Generated Corruptions -- An EndoVis'24 Challenge
SegSTRONG-C provides a new benchmark where top models reach 0.9394 DSC and 0.9301 NSD on corrupted surgical tool segmentation tests, showing conventional techniques help but calling for more innovative robustness methods.