SOTAVerified

Zero Shot Segmentation

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 134 papers

TitleStatusHype
Unleashing the Potential of SAM2 for Biomedical Images and Videos: A SurveyCode5
Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object DetectionCode5
Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive ReinforcementCode4
ZIM: Zero-Shot Image Matting for AnythingCode3
Zero-Shot Surgical Tool Segmentation in Monocular Video Using Segment Anything Model 2Code3
RobustSAM: Segment Anything Robustly on Degraded ImagesCode3
A Simple Framework for Open-Vocabulary Segmentation and DetectionCode3
Universal Instance Perception as Object Discovery and RetrievalCode3
Generalized Decoding for Pixel, Image, and LanguageCode3
CellViT++: Energy-Efficient and Adaptive Cell Segmentation and Classification Using Foundation ModelsCode2
3DGS-CD: 3D Gaussian Splatting-based Change Detection for Physical Object RearrangementCode2
Interpreting and Editing Vision-Language Representations to Mitigate HallucinationsCode2
MedCLIP-SAMv2: Towards Universal Text-Driven Medical Image SegmentationCode2
VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentationCode2
DiffCut: Catalyzing Zero-Shot Semantic Segmentation with Diffusion Features and Recursive Normalized CutCode2
Test-Time Adaptation with SaLIP: A Cascade of SAM and CLIP for Zero shot Medical Image SegmentationCode2
MedCLIP-SAM: Bridging Text and Image Towards Universal Medical Image SegmentationCode2
Diffuse, Attend, and Segment: Unsupervised Zero-Shot Segmentation using Stable DiffusionCode2
DatasetDM: Synthesizing Data with Perception Annotations Using Diffusion ModelsCode2
Hierarchical Open-vocabulary Universal Image SegmentationCode2
Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion ModelsCode2
Side Adapter Network for Open-Vocabulary Semantic SegmentationCode2
Language-driven Semantic SegmentationCode2
Compress Any Segment Anything Model (SAM)Code1
Zero-Shot Tree Detection and Segmentation from Aerial Forest ImageryCode1
COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-trainingCode1
Evaluation Study on SAM 2 for Class-agnostic Instance-level SegmentationCode1
SAM-UNet:Enhancing Zero-Shot Segmentation of SAM for Universal Medical ImagesCode1
MeshSegmenter: Zero-Shot Mesh Semantic Segmentation via Texture SynthesisCode1
Frenet-Serret Frame-based Decomposition for Part Segmentation of 3D Curvilinear StructuresCode1
TV-SAM: Increasing Zero-Shot Segmentation Performance on Multimodal Medical Images Using GPT-4 Generated Descriptive Prompts Without Human AnnotationCode1
MatSAM: Efficient Extraction of Microstructures of Materials via Visual Large ModelCode1
Spectral Prompt Tuning:Unveiling Unseen Classes for Zero-Shot Semantic SegmentationCode1
Grounding Everything: Emerging Localization Properties in Vision-Language TransformersCode1
GeoSAM: Fine-tuning SAM with Multi-Modal Prompts for Mobility Infrastructure SegmentationCode1
Learning Mask-aware CLIP Representations for Zero-Shot SegmentationCode1
MediViSTA: Medical Video Segmentation via Temporal Fusion SAM Adaptation for EchocardiographyCode1
Zero-Shot Edge Detection with SCESAME: Spectral Clustering-based Ensemble for Segment Anything Model EstimationCode1
TongueSAM: An Universal Tongue Segmentation Model Based on SAM with Zero-ShotCode1
Training-free Object Counting with PromptsCode1
How to Efficiently Adapt Large Segmentation Model(SAM) to Medical ImagesCode1
Primitive Generation and Semantic-related Alignment for Universal Zero-Shot SegmentationCode1
TomoSAM: a 3D Slicer extension using SAM for tomography segmentationCode1
PaintSeg: Training-free Segmentation via PaintingCode1
One-Prompt to Segment All Medical ImagesCode1
Segment Anything Model for Medical Images?Code1
Generalist Vision Foundation Models for Medical Imaging: A Case Study of Segment Anything Model on Zero-Shot Medical SegmentationCode1
Segment Anything Model for Medical Image Analysis: an Experimental StudyCode1
A Closer Look at the Explainability of Contrastive Language-Image Pre-trainingCode1
ZegOT: Zero-shot Segmentation Through Optimal Transport of Text PromptsCode1
Show:102550
← PrevPage 1 of 3Next →
All datasetsSegmentation in the WildADE20K training-free zero-shot segmentation

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Grounded HQ-SAMMean AP49.6—Unverified
2Grounded-SAMMean AP46—Unverified
3UNINEXTMean AP42.1—Unverified
4HIPIEMean AP41.6—Unverified
5SANMean AP41.4—Unverified
6odiseMean AP38.7—Unverified
7OpenSEEDMean AP36.1—Unverified
8OpenSDMean AP35.8—Unverified
9SGinW_Team (X-Decoder-L)Mean AP32.2—Unverified
10SGinW_Team (X-Decoder-B)Mean AP27.7—Unverified
#ModelMetricClaimedVerifiedStatus
1COSMOS ViT-B/16mIoU17.7—Unverified
2GEM (MetaCLIP)mIoU17.1—Unverified
3GEM (CLIP)mIoU15.7—Unverified
4CLIPSurgerymIoU12.9—Unverified
5MaskCLIPmIoU10.2—Unverified