ICCV 2025 accepted papers on Detection and segmentation
220 of the 2,701 papers accepted at ICCV 2025 (8.1%) are on Detection and segmentation, compared with 12.1% at ICCV 2023. They fall into 5 topics.
See ICCV 2025 in the interactive map
Submitting to ICCV? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Open-set and few-shot detection · 111 papers
- YOLOE: Real-Time Seeing Anything (185 citations)
- RadGPT: Constructing 3D Image-Text Tumor Datasets (40 citations)
- Harnessing Vision Foundation Models for High-Performance, Training-Free Open Vocabulary Segmentation (34 citations)
- CorrCLIP: Reconstructing Patch Correlations in CLIP for Open-Vocabulary Semantic Segmentation (26 citations)
- OmniSAM: Omnidirectional Segment Anything Model for UDA in Panoramic Semantic Segmentation (23 citations)
- ObjectRelator: Enabling Cross-View Object Relation Understanding Across Ego-Centric and Exo-Centric Perspectives (19 citations)
- HiMTok: Learning Hierarchical Mask Tokens for Image Segmentation with Large Multimodal Model (18 citations)
- CLIPer: Hierarchically Improving Spatial Representation of CLIP for Open-Vocabulary Semantic Segmentation (15 citations)
- OpenRSD: Towards Open-prompts for Object Detection in Remote Sensing Images (15 citations)
- Triad: Empowering LMM-based Anomaly Detection with Expert-guided Region-of-Interest Tokenizer and Manufacturing Process (15 citations)
- DeRIS: Decoupling Perception and Cognition for Enhanced Referring Image Segmentation through Loopback Synergy (14 citations)
- MaskSAM: Auto-prompt SAM with Mask Classification for Volumetric Medical Image Segmentation (14 citations)
and 99 more in the interactive map.
CNN features · 52 papers
- WaveMamba: Wavelet-Driven Mamba Fusion for RGB-Infrared Object Detection (59 citations)
- Rethinking Multi-modal Object Detection from the Perspective of Mono-Modality Feature Learning (42 citations)
- TinyViM: Frequency Decoupling for Tiny Hybrid Vision Mamba (42 citations)
- Harnessing Massive Satellite Imagery with Efficient Masked Image Modeling (31 citations)
- Dynamic Dictionary Learning for Remote Sensing Image Segmentation (28 citations)
- Frequency-Dynamic Attention Modulation For Dense Prediction (25 citations)
- Event-based Tiny Object Detection: A Benchmark Dataset and Baseline (24 citations)
- ESCNet:Edge-Semantic Collaborative Network for Camouflaged Object Detection (22 citations)
- Rectifying Magnitude Neglect in Linear Attention (22 citations)
- Uncertainty-Aware Gradient Stabilization for Small Object Detection (19 citations)
- Improving SAM for Camouflaged Object Detection via Dual Stream Adapters (16 citations)
- Revisiting Efficient Semantic Segmentation: Learning Offsets for Better Spatial and Class Feature Alignment (14 citations)
and 40 more in the interactive map.
3D detection and driving · 30 papers
- Height-Fidelity Dense Global Fusion for Multi-modal 3D Object Detection (14 citations)
- R-LiViT: A LiDAR-Visual-Thermal Dataset Enabling Vulnerable Road User Focused Roadside Perception (14 citations)
- CVFusion: Cross-View Fusion of 4D Radar and Camera for 3D Object Detection (13 citations)
- Detect Anything 3D in the Wild (12 citations)
- Doppler-Aware LiDAR-RADAR Fusion for Weather-Robust 3D Detection (12 citations)
- Harnessing Uncertainty-aware Bounding Boxes for Unsupervised 3D Object Detection (11 citations)
- 3D-MOOD: Lifting 2D to 3D for Monocular Open-Set Object Detection (9 citations)
- EVT: Efficient View Transformation for Multi-Modal 3D Object Detection (8 citations)
- RCTDistill: Cross-Modal Knowledge Distillation Framework for Radar-Camera 3D Object Detection with Temporal Fusion (8 citations)
- Self-Supervised Sparse Sensor Fusion for Long Range Perception (6 citations)
- OcRFDet: Object-Centric Radiance Fields for Multi-View 3D Object Detection in Autonomous Driving (5 citations)
- OpenM3D: Open Vocabulary Multi-view Indoor 3D Object Detection without Human Annotations (5 citations)
and 18 more in the interactive map.
Weakly supervised detection and segmentation · 14 papers
- From Easy to Hard: Progressive Active Learning Framework for Infrared Small Target Detection with Single Point Supervision (19 citations)
- Highlight What You Want: Weakly-Supervised Instance-Level Controllable Infrared-Visible Image Fusion (13 citations)
- RIPE: Reinforcement Learning on Unlabeled Image Pairs for Robust Keypoint Extraction (11 citations)
- STEP-DETR: Advancing DETR-based Semi-Supervised Object Detection with Super Teacher and Pseudo-Label Guided Text Queries (6 citations)
- Beyond Single Images: Retrieval Self-Augmented Unsupervised Camouflaged Object Detection (5 citations)
- CoralSRT: Revisiting Coral Reef Semantic Segmentation by Feature Rectification via Self-supervised Guidance (4 citations)
- Know Your Attention Maps: Class-specific Token Masking for Weakly Supervised Semantic Segmentation (4 citations)
- Class Token as Proxy: Optimal Transport-assisted Proxy Learning for Weakly Supervised Semantic Segmentation (3 citations)
- Selective Contrastive Learning for Weakly Supervised Affordance Grounding (3 citations)
- Bias-Resilient Weakly Supervised Semantic Segmentation Using Normalizing Flows (2 citations)
- Power of Cooperative Supervision: Multiple Teachers Framework for Advanced 3D Semi-Supervised Object Detection (2 citations)
- LLM-Assisted Semantic Guidance for Sparsely Annotated Remote Sensing Object Detection (1 citation)
and 2 more in the interactive map.
Tracking · 13 papers
- TAPNext: Tracking Any Point (TAP) as Next Token Prediction (31 citations)
- Efficient Track Anything (25 citations)
- LA-MOTR: End-to-End Multi-Object Tracking by Learnable Association (13 citations)
- CoopTrack: Exploring End-to-End Learning for Efficient Cooperative Sequential Perception (11 citations)
- PropVG: End-to-End Proposal-Driven Visual Grounding with Multi-Granularity Discrimination (9 citations)
- Multi-Object Sketch Animation by Scene Decomposition and Motion Planning (6 citations)
- VOVTrack: Exploring the Potentiality in Raw Videos for Open-Vocabulary Multi-Object Tracking (5 citations)
- Language Decoupling with Fine-grained Knowledge Guidance for Referring Multi-object Tracking (3 citations)
- Online Dense Point Tracking with Streaming Memory (3 citations)
- COVTrack: Continuous Open-Vocabulary Tracking via Adaptive Multi-Cue Fusion (2 citations)
- Tracking Tiny Drones against Clutter: Large-Scale Infrared Benchmark with Motion-Centric Adaptive Algorithm (2 citations)
- Attention to Trajectory: Trajectory-Aware Open-Vocabulary Tracking (1 citation)
and 1 more in the interactive map.
