CVPR 2025 accepted papers on Detection and segmentation

Open-set and few-shot detection · 99 papers

• 5%>100%: Breaking Performance Shackles of Full Fine-Tuning on Visual Recognition Tasks (70 citations)
• SegEarth-OV: Towards Training-Free Open-Vocabulary Segmentation for Remote Sensing Images (51 citations)
• Forensics Adapter: Adapting CLIP for Generalizable Face Forgery Detection (46 citations)
• LLMDet: Learning Strong Open-Vocabulary Object Detectors under the Supervision of Large Language Models (41 citations)
• AeroGen: Enhancing Remote Sensing Object Detection with Diffusion-Driven Data Generation (33 citations)
• Segmenting Maxillofacial Structures in CBCT Volumes (30 citations)
• ROS-SAM: High-Quality Interactive Segmentation for Remote Sensing Moving Object (25 citations)
• MANTA: A Large-Scale Multi-View and Visual-Text Anomaly Detection Dataset for Tiny Objects (19 citations)
• Mask-Adapter: The Devil is in the Masks for Open-Vocabulary Segmentation (18 citations)
• Interactive Medical Image Segmentation: A Benchmark Dataset and Baseline (17 citations)
• DeCLIP: Decoupled Learning for Open-Vocabulary Dense Perception (14 citations)
• Show and Segment: Universal Medical Image Segmentation via In-Context Learning (14 citations)

and 87 more in the interactive map.

CNN features · 44 papers

• MambaVision: A Hybrid Mamba-Transformer Vision Backbone (308 citations)
• DEIM: DETR with Improved Matching for Fast Convergence (273 citations)
• OverLoCK: An Overview-first-Look-Closely-next ConvNet with Context-Mixing Dynamic Kernels (93 citations)
• SegMAN: Omni-scale Context Modeling with State Space Models and Local Attention for Semantic Segmentation (65 citations)
• SCSegamba: Lightweight Structure-Aware Vision Mamba for Crack Segmentation in Structures (64 citations)
• Task-driven Image Fusion with Learnable Fusion Loss (46 citations)
• Golden Cudgel Network for Real-Time Semantic Segmentation (41 citations)
• DefMamba: Deformable Visual State Space Model (37 citations)
• nnWNet: Rethinking the Use of Transformers in Biomedical Image Segmentation and Calling for a Unified Evaluation Benchmark (36 citations)
• SET: Spectral Enhancement for Tiny Object Detection (28 citations)
• Wavelet and Prototype Augmented Query-based Transformer for Pixel-level Surface Defect Detection (23 citations)
• GroupMamba: Efficient Group-Based Visual State Space Model (22 citations)

and 32 more in the interactive map.

3D detection and driving · 26 papers

• MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (25 citations)
• V2X-R: Cooperative LiDAR-4D Radar Fusion with Denoising Diffusion for 3D Object Detection (24 citations)
• RaCFormer: Towards High-Quality 3D Object Detection via Query-based Radar-Camera Fusion (19 citations)
• Resilient Sensor Fusion Under Adverse Sensor Failures via Multi-Modal Expert Fusion (12 citations)
• Cubify Anything: Scaling Indoor 3D Object Detection (10 citations)
• SparseAlign: a Fully Sparse Framework for Cooperative Object Detection (10 citations)
• ViKIENet: Towards Efficient 3D Object Detection with Virtual Key Instance Enhanced Network (10 citations)
• Hyperdimensional Uncertainty Quantification for Multimodal Uncertainty Fusion in Autonomous Vehicles Perception (9 citations)
• InstanceGaussian: Appearance-Semantic Joint Gaussian Representation for 3D Instance-Level Perception (9 citations)
• BEVDiffuser: Plug-and-Play Diffusion Model for BEV Denoising with Ground-Truth Guidance (8 citations)
• Learning to Detect Objects from Multi-Agent LiDAR Scans without Manual Labels (8 citations)
• PillarHist: A Quantization-aware Pillar Feature Encoder based on Height-aware Histogram (8 citations)

and 14 more in the interactive map.

Weakly supervised detection and segmentation · 17 papers

• RSAR: Restricted State Angle Resolver and Rotated SAR Benchmark (19 citations)
• FFR: Frequency Feature Rectification for Weakly Supervised Semantic Segmentation (15 citations)
• Point2RBox-v2: Rethinking Point-supervised Oriented Object Detection with Spatial Layout Among Instances (14 citations)
• POT: Prototypical Optimal Transport for Weakly Supervised Semantic Segmentation (14 citations)
• Prototype-Based Image Prompting for Weakly Supervised Histopathological Image Segmentation (14 citations)
• Weakly Supervised Semantic Segmentation via Progressive Confidence Region Expansion (14 citations)
• Multi-Label Prototype Visual Spatial Search for Weakly Supervised Semantic Segmentation (6 citations)
• Sketchy Bounding-box Supervision for 3D Instance Segmentation (6 citations)
• WeakMCN: Multi-task Collaborative Network for Weakly Supervised Referring Expression Comprehension and Segmentation (6 citations)
• UCOD-DPL: Unsupervised Camouflaged Object Detection via Dynamic Pseudo-label Learning (5 citations)
• ABBSPO: Adaptive Bounding Box Scaling and Symmetric Prior based Orientation Prediction for Detecting Aerial Image Objects (4 citations)
• CSC-PA: Cross-image Semantic Correlation via Prototype Attentions for Single-network Semi-supervised Breast Tumor Segmentation (4 citations)

and 5 more in the interactive map.

Tracking · 12 papers

• Multiple Object Tracking as ID Prediction (61 citations)
• A Distractor-Aware Memory for Visual Object Tracking with SAM2 (42 citations)
• Focusing on Tracks for Online Multi-Object Tracking (31 citations)
• MUST: The First Dataset and Unified Framework for Multispectral UAV Single Object Tracking (14 citations)
• Autoregressive Sequential Pretraining for Visual Tracking (11 citations)
• DreamTrack: Dreaming the Future for Multimodal Visual Object Tracking (11 citations)
• Omnidirectional Multi-Object Tracking (7 citations)
• CholecTrack20: A Multi-Perspective Tracking Dataset for Surgical Tools (4 citations)
• Descriptor-In-Pixel : Point-Feature Tracking For Pixel Processor Arrays (4 citations)
• Exploring Temporally-Aware Features for Point Tracking (2 citations)
• GRAE-3DMOT: Geometry Relation-Aware Encoder for Online 3D Multi-Object Tracking (1 citation)
• Tracktention: Leveraging Point Tracking to Attend Videos Faster and Better