SAM2S: Segment Anything in Surgical Videos via Semantic Long-term Tracking

Video Demonstrations

SAM2S achieves real-time performance at 68 FPS with robust long-term tracking across diverse surgical procedures

CIS-Test Dataset: Long-term tracking in cholecystectomy

EndoVis18 Dataset: Zero-shot generalization on unseen nephrectomy procedures

RARP50 Dataset: Robust segmentation in robot-assisted radical prostatectomy

Endoscapes2023 Dataset: Robust tissue tracking under viewpoint changes and tissue deformation on unseen video

NUH In-House Dataset: Generalization to dynamic scenes in external hysterectomy video

Abstract

Surgical video segmentation is crucial for computer-assisted surgery, enabling precise localization and tracking of instruments and tissues. Interactive Video Object Segmentation (iVOS) models such as Segment Anything Model 2 (SAM2) provide prompt-based flexibility beyond methods with predefined categories, but face challenges in surgical scenarios due to the domain gap and limited long-term tracking. To address these limitations, we construct SA-SV, the largest surgical iVOS benchmark with instance-level spatio-temporal annotations (masklets) spanning eight procedure types (61k frames, 1.6k masklets), enabling comprehensive development and evaluation for long-term tracking and zero-shot generalization. Building on SA-SV, we propose SAM2S, a foundation model enhancing SAM2 for Surgical iVOS through: (1) DiveMem, a trainable diverse memory mechanism for robust long-term tracking; (2) Temporal Semantic Learning (TSL) for instrument understanding; and (3) Ambiguity-Resilient Learning (ARL) to mitigate annotation inconsistencies across multi-source datasets. Extensive experiments demonstrate that fine-tuning on SA-SV enables substantial performance gains, with SAM2 improving by 12.99 average J&F over vanilla SAM2. SAM2S further advances performance to 80.42 average J&F, surpassing vanilla and fine-tuned SAM2 by 17.10 and 4.11 points respectively, while maintaining 68 FPS real-time inference and strong zero-shot generalization.

SA-SV Benchmark

The largest surgical iVOS benchmark with comprehensive segmentation annotations

572

Videos

61K

Frames

1.6K

Masklets

8

Procedure Types

Covered Surgical Procedures
                Cholecystectomy: Endoscapes, CholecSeg8k, CholecInstanceSeg (CIS)
Colonoscopy: PolypGen, Kvasir-SEG, BKAI-IGH, CVC-ClinicDB
Gynecology: SurgAI3.8k
Hysterectomy: AutoLaparo, ART-Net, Hyst-YT
Myotomy: DSAD
Nephrectomy: EndoVis17, EndoVis18
Prostatectomy: GraSP, RARP50
Multi-procedural: RoboTool

            

Proposed Method

Overview of SAM2S framework integrating DiveMem, TSL, and ARL for robust surgical video segmentation

Key Innovations

DiveMem: Diverse Memory for Long-term Tracking

A trainable diverse memory mechanism that employs hybrid temporal sampling during training and diversity-based frame selection during inference. DiveMem addresses viewpoint overfitting in long-term surgical tracking by maintaining both diverse long-term memory and complete short-term memory, ensuring comprehensive temporal context that training-free approaches fail to maintain.

TSL: Temporal Semantic Learning

Leverages semantic categories of surgical instruments through vision-language contrastive learning with CLIP. TSL enables semantic-aware tracking by incorporating learnable CLS tokens that attend to memory features and perform cross-attention with current frame features, while preserving class-agnostic generalization capability.

ARL: Ambiguity-Resilient Learning

Handles annotation inconsistencies across multi-source datasets through uniform label softening using Gaussian kernel convolution. ARL transforms discrete annotation spaces into continuous probability distributions, improving model calibration and robustness at ambiguous tissue boundaries while mitigating conflicting supervision signals from varying labeling standards.

Performance Highlights

Zero-shot Generalization

80.42
Average J&F (3-click)

+17.10 over vanilla SAM2
+4.11 over fine-tuned SAM2

Real-time Performance

68
FPS (Frames Per Second)

Real-time inference on A6000 GPU
Suitable for clinical deployment

Long-term Tracking

CIS-Test (≈30 min): 89.65 J&F (+9.56)
RARP50 (325s): 79.47 J&F (+2.96)
Hyst-YT (329s): 87.46 J&F (+3.57)

Cross-procedure Generalization

EndoVis17 (unseen): 86.72 J&F
EndoVis18-I (unseen): 82.37 J&F
Strong performance on nephrectomy without training data

Note: All test subsets remain completely unseen during training, ensuring rigorous evaluation of zero-shot generalization across diverse surgical procedures.

Related Work

Explore our other works on surgical video segmentation with SAM2

SAM2 Series for Surgical Applications: Our research group has developed a comprehensive suite of approaches to adapt Segment Anything Model 2 (SAM2) for surgical video understanding. Surgical SAM 2 achieves real-time performance (86 FPS) through efficient frame pruning, ReSurgSAM2 introduces text-guided referring segmentation with credible tracking initialization for language-driven surgical scene understanding, and SAM2S (this work) provides the most generalizable solution with semantic long-term tracking and diverse memory mechanisms across diverse surgical procedures.

Acknowledgement

We would like to thank National University Hospital (NUH) for providing the external hysterectomy videos. These resources were valuable for verifying the effectiveness and robustness of our model in real-world surgical scenarios.

Citation

@article{liu2025sam2s,
  title={SAM2S: Segment Anything in Surgical Videos via Semantic Long-term Tracking},
  author={Liu, Haofeng and Wang, Ziyue and Mishra, Sudhanshu and Gao, Mingqi and
          Qin, Guanyi and Low, Chang Han and Kong, Alex Y. W. and Jin, Yueming},
  journal={arXiv preprint arXiv:2511.16618},
  year={2025}
}