Image de couverture de livre

The SAM2-to-SAM3 Gap in the Segment Anything Model Family: Why Prompt-Based Expertise Fails in Concept-Driven Image Segmentation

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Sapkota, Ranjan, Roumeliotis, Konstantinos I., Karkee, Manoj
Format:	Preprint
Publié:	2025
Sujets:	Computer Vision and Pattern Recognition Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Generalization vs. Specialization: Evaluating Segment Anything Model (SAM3) Zero-Shot Segmentation Against Fine-Tuned YOLO Detectors
par: Sapkota, Ranjan, et autres
Publié: (2025)

Vision-Language-Action (VLA) Models: Concepts, Progress, Applications and Challenges
par: Sapkota, Ranjan, et autres
Publié: (2025)

YOLOE-26: Integrating YOLO26 with YOLOE for Real-Time Open-Vocabulary Instance Segmentation
par: Sapkota, Ranjan, et autres
Publié: (2026)

Integrating YOLO11 and Convolution Block Attention Module for Multi-Season Segmentation of Tree Trunks and Branches in Commercial Apple Orchards
par: Sapkota, Ranjan, et autres
Publié: (2024)

A Review of 3D Object Detection with Vision-Language Models
par: Sapkota, Ranjan, et autres
Publié: (2025)

Generative AI in Agriculture: Creating Image Datasets Using DALL.E's Advanced Large Language Model Capabilities
par: Sapkota, Ranjan, et autres
Publié: (2023)

SAM 3: Segment Anything with Concepts
par: Carion, Nicolas, et autres
Publié: (2025)

Zero-Shot Automatic Annotation and Instance Segmentation using LLM-Generated Datasets: Eliminating Field Imaging and Manual Annotation for Deep Learning Model Development
par: Sapkota, Ranjan, et autres
Publié: (2024)

Plant Disease Detection through Multimodal Large Language Models and Convolutional Neural Networks
par: Roumeliotis, Konstantinos I., et autres
Publié: (2025)

YOLO11 and Vision Transformers based 3D Pose Estimation of Immature Green Fruits in Commercial Apple Orchards for Robotic Thinning
par: Sapkota, Ranjan, et autres
Publié: (2024)

Comparing YOLOv11 and YOLOv8 for instance segmentation of occluded and non-occluded immature green fruits in complex orchard environment
par: Sapkota, Ranjan, et autres
Publié: (2024)

Object Detection with Multimodal Large Vision-Language Models: An In-depth Review
par: Sapkota, Ranjan, et autres
Publié: (2025)

Ultralytics YOLO Evolution: An Overview of YOLO26, YOLO11, YOLOv8 and YOLOv5 Object Detectors for Computer Vision and Pattern Recognition
par: Sapkota, Ranjan, et autres
Publié: (2025)

Improved YOLOv12 with LLM-Generated Synthetic Data for Enhanced Apple Detection and Benchmarking Against YOLOv11 and YOLOv10
par: Sapkota, Ranjan, et autres
Publié: (2025)

MedSAM3: Delving into Segment Anything with Medical Concepts
par: Liu, Anglin, et autres
Publié: (2025)

PP-SAM: Perturbed Prompts for Robust Adaptation of Segment Anything Model for Polyp Segmentation
par: Rahman, Md Mostafijur, et autres
Publié: (2024)

DeiSAM: Segment Anything with Deictic Prompting
par: Shindo, Hikaru, et autres
Publié: (2024)

SAM3-I: Segment Anything with Instructions
par: Li, Jingjing, et autres
Publié: (2025)

Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentation
par: Wu, Junde, et autres
Publié: (2023)

From SAM to SAM 2: Exploring Improvements in Meta's Segment Anything Model
par: Geetha, Athulya Sundaresan, et autres
Publié: (2024)

I-MedSAM: Implicit Medical Image Segmentation with Segment Anything
par: Wei, Xiaobao, et autres
Publié: (2023)

Comparing YOLOv8 and Mask R-CNN for instance segmentation in complex orchard environments
par: Sapkota, Ranjan, et autres
Publié: (2023)

EP-SAM: Weakly Supervised Histopathology Segmentation via Enhanced Prompt with Segment Anything
par: Song, Joonhyeon, et autres
Publié: (2024)

Det-SAM2:Technical Report on the Self-Prompting Segmentation Framework Based on Segment Anything Model 2
par: Wang, Zhiting, et autres
Publié: (2024)

AM-SAM: Automated Prompting and Mask Calibration for Segment Anything Model
par: Li, Yuchen, et autres
Publié: (2024)

S-SAM: SVD-based Fine-Tuning of Segment Anything Model for Medical Image Segmentation
par: Paranjape, Jay N., et autres
Publié: (2024)

GoodSAM++: Bridging Domain and Capacity Gaps via Segment Anything Model for Panoramic Semantic Segmentation
par: Zhang, Weiming, et autres
Publié: (2024)

Hi-SAM: Marrying Segment Anything Model for Hierarchical Text Segmentation
par: Ye, Maoyuan, et autres
Publié: (2024)

Compress Any Segment Anything Model (SAM)
par: Fan, Juntong, et autres
Publié: (2025)

PaveSAM Segment Anything for Pavement Distress
par: Owor, Neema Jakisa, et autres
Publié: (2024)

Biomedical SAM 2: Segment Anything in Biomedical Images and Videos
par: Yan, Zhiling, et autres
Publié: (2024)

SAM 2: Segment Anything in Images and Videos
par: Ravi, Nikhila, et autres
Publié: (2024)

SAM-PD: How Far Can SAM Take Us in Tracking and Segmenting Anything in Videos by Prompt Denoising
par: Zhou, Tao, et autres
Publié: (2024)

Self-Prompt SAM: Medical Image Segmentation via Automatic Prompt SAM Adaptation
par: Xie, Bin, et autres
Publié: (2025)

SAM3-UNet: Simplified Adaptation of Segment Anything Model 3
par: Xiong, Xinyu, et autres
Publié: (2025)

SGP-SAM: Self-Gated Prompting for Transferring 3D Segment Anything Models to Lesion Segmentation
par: Tang, Zixuan, et autres
Publié: (2026)

Inspiring the Next Generation of Segment Anything Models: Comprehensively Evaluate SAM and SAM 2 with Diverse Prompts Towards Context-Dependent Concepts under Different Scenes
par: Zhao, Xiaoqi, et autres
Publié: (2024)

X-SAM: From Segment Anything to Any Segmentation
par: Wang, Hao, et autres
Publié: (2025)

SAM-REF: Introducing Image-Prompt Synergy during Interaction for Detail Enhancement in the Segment Anything Model
par: Yu, Chongkai, et autres
Publié: (2024)

SAM3D: Segment Anything Model in Volumetric Medical Images
par: Bui, Nhat-Tan, et autres
Publié: (2023)