Segment Anything Is Not Always Perfect: An Investigation of SAM on Different Real-world Applications

Abstract

Recently, Meta AI Research approaches a general, promptable Segment Anything Model (SAM) pre-trained on an unprecedentedly large segmentation dataset (SA-1B). Without a doubt, the emergence of SAM will yield significant benefits for a wide array of practical image segmentation applications. In this study, we conduct a series of intriguing investigations into the performance of SAM across various applications, particularly in the fields of natural images, agriculture, manufacturing, remote sensing, and healthcare. We analyze and discuss the benefits and limitations of SAM, while also presenting an outlook on its future development in segmentation tasks. By doing so, we aim to give a comprehensive understanding of SAM's practical applications. This work is expected to provide insights that facilitate future research activities toward generic segmentation. Source code is publicly available.

CVPRW Oral

References in corpus (8)

• Learning Transferable Visual Models From Natural Language Supervision
• Segment Anything Model for Medical Image Analysis: an Experimental Study
• Deep Gradient Learning for Efficient Camouflaged Object Detection
• Feature Aggregation and Propagation Network for Camouflaged Object Detection
• Video Polyp Segmentation: A Deep Learning Perspective
• Advances in Deep Concealed Scene Understanding
• SAM Struggles in Concealed Scenes -- Empirical Study on Segment Anything
• Promoting Saliency From Depth: Deep Unsupervised RGB-D Saliency Detection

Cited by in corpus (15)

• Segment Anything Model for Medical Images?
• Adapting Segment Anything Model for Change Detection in HR Remote Sensing Images
• 3DSAM-adapter: Holistic adaptation of SAM from 2D to 3D for promptable tumor segmentation
• A Survey of Sample-Efficient Deep Learning for Change Detection in Remote Sensing: Tasks, Strategies, and Challenges
• A Survey on Continual Semantic Segmentation: Theory, Challenge, Method and Application
• Deep learning based infrared small object segmentation: Challenges and future directions
• Efficient Long-Short Temporal Attention Network for Unsupervised Video Object Segmentation
• Robust Saliency-Aware Distillation for Few-shot Fine-grained Visual Recognition
• Learning Spectral-Decomposed Tokens for Domain Generalized Semantic Segmentation
• SinkSAM-Net: Knowledge-Driven Self-Supervised Sinkhole Segmentation Using Topographic Priors and Segment Anything Model
• Weakly Supervised Camouflaged Object Detection Based on the SAM Model and Mask Guidance
• 3D Object Detection and High-Resolution Traffic Parameters Extraction Using Low-Resolution LiDAR Data
• POPCat: Propagation of particles for complex annotation tasks
• A Point-Neighborhood Learning Framework for Nasal Endoscope Image Segmentation
• OBSeg: Accurate and Fast Instance Segmentation Framework Using Segmentation Foundation Models with Oriented Bounding Box Prompts