本次CVPR25涉及到语义分割的文章大约有144篇，粗略分为以下几类：其中与医学图像相关的占比是最多的，值得注意的是开放词汇语义分割今年也有不少。

探索类

1. Your ViT is Secretly an Image Segmentation Model

2. Convex Combination Star Shape Prior for Data-driven Image Semantic Segmentation

3. Advancing Manga Analysis: Comprehensive Segmentation Annotations for the Manga109 Dataset

4. FisherTune: Fisher-Guided Robust Tuning of Vision Foundation Models for Domain Generalized Segmentation

5. Scaling up Image Segmentation across Data and Tasks

6. NightAdapter: Learning a Frequency Adapter for Generalizable Night-time Scene Segmentation

7. SegMAN: Omni-scale Context Modeling with State Space Models and Local Attention for Semantic Segmentation

8. Universal Domain Adaptation for Semantic Segmentation

9. The Impact Label Noise and Choice of Threshold has on Cross-Entropy and Soft-Dice in Image Segmentation

10. UNICL-SAM: Uncertainty-Driven In-Context Segmentation with Part Prototype Discovery

实时语义分割

1. Golden Cudgel Network for Real-Time Semantic Segmentation

高分辨率语义分割

1. Boosting the Dual-Stream Architecture in Ultra-High Resolution Segmentation with Resolution-Biased Uncertainty Estimatio

少样本语义分割

1. Text Augmented Correlation Transformer For Few-shot Classification & Segmentation

2. DSV-LFS: Unifying LLM-Driven Semantic Cues with Visual Features for Robust Few-Shot Segmentation

3. The Devil is in Low-Level Features for Cross-Domain Few-Shot Segmentation

4. Dual-Agent Optimization framework for Cross-Domain Few-Shot Segmentation

弱监督语义分割

1. Exploring CLIP's Dense Knowledge for Weakly Supervised Semantic Segmentation

2. Prototype-Based Image Prompting for Weakly Supervised Histopathological Image Segmentation

3. Weakly Supervised Semantic Segmentation via Progressive Confidence Region Expansion

4. WISH: Weakly Supervised Instance Segmentation using Heterogeneous Labels

5. FFR: Frequency Feature Rectification for Weakly Supervised Semantic Segmentation

6. POT: Prototypical Optimal Transport for Weakly Supervised Semantic Segmentation

7. Multi-Label Prototype Visual Spatial Search for Weakly Supervised Semantic Segmentation

8. Soft Self-labeling and Potts Relaxations for Weakly-supervised Segmentation

半监督语义分割

1. Improving Semi-Supervised Semantic Segmentation with Sliced-Wasserstein Feature Alignment and Uniformity

2. SemiDAViL: Semi-supervised Domain Adaptation with Vision-Language Guidance for Semantic Segmentation

开放词汇语义分割

1. Mask-Adapter: The Devil is in the Masks for Open-Vocabulary Segmentation

2. LPOSS: Label Propagation Over Patches and Pixels for Open-vocabulary Semantic Segmentation

3. Understanding Fine-tuning CLIP for Open-vocabulary Semantic Segmentation in Hyperbolic Space

4. Semantic Library Adaptation: LoRA Retrieval and Fusion for Open-Vocabulary Semantic Segmentation

5. Dual Semantic Guidance for Open Vocabulary Semantic Segmentation

6. Effective SAM Combination for Open-Vocabulary Semantic Segmentation

7. Exploring Simple Open-Vocabulary Semantic Segmentation

8. Parameter-efficient Fine-tuning in Hyperspherical Space for Open-vocabulary Semantic Segmentation

9. Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic Segmentation

10. DPSeg: Dual-Prompt Cost Volume Learning for Open-Vocabulary Semantic Segmentation

全景分割

1. Zero-Shot 4D Lidar Panoptic Segmentation

2. Scene-Centric Unsupervised Panoptic Segmentation

持续学习语义分割

1. Beyond Background Shift: Rethinking Instance Replay in Continual Semantic Segmentation

2. Rethinking Query-based Transformer for Continual Image Segmentation

3. Towards Continual Universal Segmentation

类别增量语义分割

1. CoMBO: Conflict Mitigation via Branched Optimization for Class Incremental Segmentation

医学图像语义分割

1. Steady Progress Beats Stagnation: Mutual Aid of Foundation and Conventional Models in Mixed Domain Semi-Supervised Medical Image Segmentation

2. Unified Medical Lesion Segmentation via Self-referring Indicator

3. Revisiting MAE Pre-training for 3D Medical Image Segmentation

4. Test-Time Domain Generalization via Universe Learning: A Multi-Graph Matching Approach for Medical Image Segmentation

5. Rethinking Decoder Design: Improving Biomarker Segmentation Using Depth-to-Space Restoration and Residual Linear Attention

6. Show and Segment: Universal Medical Image Segmentation via In-Context Learning

7. A Semantic Knowledge Complementarity based Decoupling Framework for Semi-supervised Class-imbalanced Medical Image Segmentation

8. vesselFM: A Foundation Model for Universal 3D Blood Vessel Segmentation

9. Advancing Generalizable Tumor Segmentation with Anomaly-Aware Open-Vocabulary Attention Maps and Frozen Foundation Diffusion Models

10. Interactive Medical Image Segmentation: A Benchmark Dataset and Baseline

11. Cross-Modal Interactive Perception Network with Mamba for Lung Tumor Segmentation in PET-CT Images

12. SuperLightNet: Lightweight Parameter Aggregation Network for Multimodal Brain Tumor Segmentation

13. CSC-PA: Cross-image Semantic Correlation via Prototype Attentions for Single-network Semi-supervised Breast Tumor Segmentation

14. LesionLocator: Zero-Shot Universal Tumor Segmentation and Tracking in 3D Whole-Body Imaging

15. Noise-Consistent Siamese-Diffusion for Medical Image Synthesis and Segmentation

16. beta-FFT: Nonlinear Interpolation and Differentiated Training Strategies for Semi-Supervised Medical Image Segmentation

17. EffiDec3D: An Optimized Decoder for High-Performance and Efficient 3D Medical Image Segmentation

18. KMD: Koopman Multi-modality Decomposition for Generalized Brain Tumor Segmentation under Incomplete Modalities

19. Annotation Ambiguity Aware Semi-Supervised Medical Image Segmentation

20. Enhancing SAM with Efficient Prompting and Preference Optimization for Semi-supervised Medical Image Segmentation

21. Minding Fuzzy Regions: A Data-driven Alternating Learning Paradigm for Stable Lesion Segmentation

22. Learning Dynamic Collaborative Network for Semi-supervised 3D Vessel Segmentation

23. nnWNet: Rethinking the Use of Transformers in Biomedical Image Segmentation and Calling for a Unified Evaluation Benchmark

24. Boost the Inference with Co-training: A Depth-guided Mutual Learning Framework for Semi-supervised Medical Polyp Segmentation

25. Incomplete Multi-modal Brain Tumor Segmentation via Learnable Sorting State Space Model

Amodal 语义分割

1. Towards Efficient Foundation Model for Zero-shot Amodal Segmentation

2. EntityErasure: Erasing Entity Cleanly via Amodal Entity Segmentation and Completion

Part语义分割

1. Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentation

2. CALICO: Part-Focused Semantic Co-Segmentation with Large Vision-Language Models

遥感场景语义分割

1. Exact: Exploring Space-Time Perceptive Clues for Weakly Supervised Satellite Image Time Series Semantic Segmentation

2. SegEarth-OV: Towards Training-Free Open-Vocabulary Segmentation for Remote Sensing Images

3. ROS-SAM: High-Quality Interactive Segmentation for Remote Sensing Moving Object

裂缝场景语义分割

1. SCSegamba: Lightweight Structure-Aware Vision Mamba for Crack Segmentation in Structures

视频语义分割

1. Semantic and Sequential Alignment for Referring Video Object Segmentation

2. VidSeg: Training-free Video Semantic Segmentation based on Diffusion Models

3. ViCaS: A Dataset for Combining Holistic and Pixel-level Video Understanding using Captions with Grounded Segmentation

4. M^3-VOS: Multi-Phase, Multi-Transition, and Multi-Scenery Video Object Segmentation

5. The Devil is in Temporal Token: High Quality Video Reasoning Segmentation

6. High Temporal Consistency through Semantic Similarity Propagation in Semi-Supervised Video Semantic Segmentation for Autonomous Flight

7. Exploiting Temporal State Space Sharing for Video Semantic Segmentation

8. SAM-I2V: Upgrading SAM to Support Promptable Video Segmentation with Less than 0.2% Training Cost

9. AMWISE: Infusing Wisdom in SAM2 for Text-Driven Video Segmentation

10. DeNVeR: Deformable Neural Vessel Representations for Unsupervised Video Vessel Segmentation

11. Using Diffusion Priors for Video Amodal Segmentation

12. GLUS: Global-Local Reasoning Unified into A Single Large Language Model for Video Segmentation

13. LiVOS: Light Video Object Segmentation with Gated Linear Matching

14. Decoupled Motion Expression Video Segmentation

RGB-X语义分割

1. DFormerv2: Geometry Self-Attention for RGBD Semantic Segmentation

2. Keep the Balance: A Parameter-Efficient Symmetrical Framework for RGB+X Semantic Segmentation

BEV语义分割

1. Generative Map Priors for Collaborative BEV Semantic Segmentation

语义分割新任务

1. WeakMCN: Multi-task Collaborative Network for Weakly Supervised Referring Expression Comprehension and Segmentation

2. MaSS13K: A Matting-level Semantic Segmentation Benchmark

3. SUM Parts: Benchmarking Part-Level Semantic Segmentation of Urban Meshes

4. A Dataset for Semantic Segmentation in the Presence of Unknowns

交互式语义分割

1. NTClick: Achieving Precise Interactive Segmentation With Noise-tolerant Clicks

2. Repurposing Stable Diffusion Attention for Training-Free Unsupervised Interactive Segmentation

指代语义分割

1. Hybrid Global-Local Representation with Augmented Spatial Guidance for Zero-Shot Referring Image Segmentation

3D场景语义分割

1. BFANet: Revisiting 3D Semantic Segmentation with Boundary Feature Analysis

2. 3D Dental Model Segmentation with Geometrical Boundary Preserving

3. Layered Motion Fusion: Lifting Motion Segmentation to 3D in Egocentric Videos

4. LogoSP: Local-global Grouping of Superpoints for Unsupervised Semantic Segmentation of 3D Point Clouds

5. Generalized Few-shot 3D Point Cloud Segmentation with Vision-Language Model

6. 3D-AVS: LiDAR-based 3D Auto-Vocabulary Segmentation

7. Generative Hard Example Augmentation for Semantic Point Cloud Segmentation

8. Hyperbolic Uncertainty-Aware Few-Shot Incremental Point Cloud Segmentation

9. CamPoint: Boosting Point Cloud Segmentation with Virtual Camera

10. OnlineAnySeg: Online Zero-Shot 3D Segmentation by Visual Foundation Model Guided 2D Mask Merging

11. D^3CTTA: Domain-Dependent Decorrelation for Continual Test-Time Adaption of 3D LiDAR Segmentation

12. Towards Explicit Geometry-Reflectance Collaboration for Generalized LiDAR Segmentation in Adverse Weather

13. PanoGS: Gaussian-based Panoptic Segmentation for 3D Open Vocabulary Scene Understanding

14. Rethinking End-to-End 2D to 3D Scene Segmentation in Gaussian Splatting

15. Relation3D : Enhancing Relation Modeling for Point Cloud Instance Segmentation

16. Functionality Understanding and Segmentation in 3D Scenes

17. An End-to-End Robust Point Cloud Semantic Segmentation Network with Single-Step Conditional Diffusion Models

18. COB-GS: Clear Object Boundaries in 3DGS Segmentation Based on Boundary-Adaptive Gaussian Splitting

19. Spotting the Unexpected (STU): A 3D LiDAR Dataset for Anomaly Segmentation in Autonomous Driving

20. Mosaic3D: Foundation Dataset and Model for Open-Vocabulary 3D Segmentation

21. Exploring Scene Affinity for Semi-Supervised LiDAR Semantic Segmentation

22. No Thing, Nothing: Highlighting Safety-Critical Classes for Robust LiDAR Semantic Segmentation in Adverse Weather

23. HiLoTs: High-Low Temporal Sensitive Representation Learning for Semi-Supervised LiDAR Segmentation in Autonomous Driving

VLM-Based语义分割

1. HyperSeg: Hybrid Segmentation Assistant with Fine-grained Visual Perceiver

2. POPEN: Preference-Based Optimization and Ensemble for LVLM-Based Reasoning Segmentation

3. Mamba as a Bridge: Where Vision Foundation Models Meet Vision Language Models for Domain-Generalized Semantic Segmentation

文档语义分割

1. DocSAM: Unified Document Image Segmentation via Query Decomposition and Heterogeneous Mixed Learning

音频-视觉联合语义分割

1. TSAM: Temporal SAM Augmented with Multimodal Prompts for Referring Audio-Visual Segmentation

2. Revisiting Audio-Visual Segmentation with Vision-Centric Transformer

3. Dynamic Derivation and Elimination: Audio Visual Segmentation with Enhanced Audio Semantics

4. Robust Audio-Visual Segmentation via Audio-Guided Visual Convergent Alignment

实例分割

1. PolarNeXt: Rethink Instance Segmentation with Polar Representation

2. Sketchy Bounding-box Supervision for 3D Instance Segmentation

3. Insightful Instance Features for 3D Instance Segmentation

4. SAM2Object: Consolidating View Consistency via SAM2 for Zero-Shot 3D Instance Segmentation

5. v-CLR: View-Consistent Learning for Open-World Instance Segmentation

6. Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking

7. RipVIS: Rip Currents Video Instance Segmentation Benchmark for Beach Monitoring and Safety

8. Minimizing Labeled, Maximizing Unlabeled: An Image-Driven Approach for Video Instance Segmentation

9. Audio-Visual Instance Segmentation

10. Foveated Instance Segmentation