CVPR 2023 accepted papers on Detection and segmentation
332 of the 2,359 papers accepted at CVPR 2023 (14.1%) are on Detection and segmentation, compared with 16.2% at CVPR 2022. They fall into 5 topics.
See CVPR 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Open-set and few-shot detection · 141 papers
- Objaverse: A Universe of Annotated 3D Objects (698 citations)
- EVA: Exploring the Limits of Masked Visual Representation Learning at Scale (478 citations)
- Open-Vocabulary Semantic Segmentation With Mask-Adapted CLIP (448 citations)
- Side Adapter Network for Open-Vocabulary Semantic Segmentation (308 citations)
- DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detection (280 citations)
- ZegCLIP: Towards Adapting CLIP for Zero-Shot Semantic Segmentation (233 citations)
- Hierarchical Fine-Grained Image Forgery Detection and Localization (200 citations)
- Cut and Learn for Unsupervised Object Detection and Instance Segmentation (185 citations)
- GRES: Generalized Referring Expression Segmentation (182 citations)
- Hierarchical Dense Correlation Distillation for Few-Shot Segmentation (167 citations)
- CoWs on Pasture: Baselines and Benchmarks for Language-Driven Zero-Shot Object Navigation (166 citations)
- CORA: Adapting CLIP for Open-Vocabulary Detection With Region Prompting and Anchor Pre-Matching (152 citations)
and 129 more in the interactive map.
CNN features · 93 papers
- YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors (11,788 citations)
- CDDFuse: Correlation-Driven Dual-Branch Feature Decomposition for Multi-Modality Image Fusion (924 citations)
- PIDNet: A Real-Time Semantic Segmentation Network Inspired by PID Controllers (735 citations)
- Mask DINO: Towards a Unified Transformer-Based Framework for Object Detection and Segmentation (543 citations)
- OneFormer: One Transformer To Rule Universal Image Segmentation (490 citations)
- Camouflaged Object Detection With Feature Decomposition and Edge Reconstruction (377 citations)
- MobileOne: An Improved One Millisecond Mobile Backbone (376 citations)
- Dense Distinct Query for End-to-End Object Detection (311 citations)
- Feature Shrinkage Pyramid for Camouflaged Object Detection With Transformers (289 citations)
- MetaFusion: Infrared and Visible Image Fusion via Meta-Feature Embedding From Object Detection (253 citations)
- Adaptive Sparse Convolutional Networks With Global Context Enhancement for Faster Object Detection on Drone Images (250 citations)
- DETRs With Hybrid Matching (245 citations)
and 81 more in the interactive map.
3D detection and driving · 47 papers
- VoxelNeXt: Fully Sparse VoxelNet for 3D Object Detection and Tracking (464 citations)
- BEVFormer v2: Adapting Modern Image Backbones to Bird's-Eye-View Recognition via Perspective Supervision (322 citations)
- V2V4Real: A Real-World Large-Scale Dataset for Vehicle-to-Vehicle Cooperative Perception (243 citations)
- Virtual Sparse Convolution for Multimodal 3D Object Detection (223 citations)
- MSMDFusion: Fusing LiDAR and Camera at Multiple Scales With Multi-Depth Seeds for 3D Object Detection (148 citations)
- LargeKernel3D: Scaling Up Kernels in 3D Sparse CNNs (135 citations)
- BEVHeight: A Robust Framework for Vision-Based Roadside 3D Object Detection (121 citations)
- Omni3D: A Large Benchmark and Model for 3D Object Detection in the Wild (106 citations)
- Habitat-Matterport 3D Semantics Dataset (89 citations)
- OcTr: Octree-Based Transformer for 3D Object Detection (86 citations)
- UniDistill: A Universal Cross-Modality Knowledge Distillation Framework for 3D Object Detection in Bird's-Eye View (75 citations)
- Neural Map Prior for Autonomous Driving (65 citations)
and 35 more in the interactive map.
Weakly supervised detection and segmentation · 34 papers
- Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentation (565 citations)
- CLIP Is Also an Efficient Segmenter: A Text-Driven Approach for Weakly Supervised Semantic Segmentation (236 citations)
- Token Contrast for Weakly-Supervised Semantic Segmentation (207 citations)
- Mapping Degeneration Meets Label Evolution: Learning Infrared Small Target Detection With Single Point Supervision (135 citations)
- Consistent-Teacher: Towards Reducing Inconsistent Pseudo-Targets in Semi-Supervised Object Detection (119 citations)
- Boundary-Enhanced Co-Training for Weakly Supervised Semantic Segmentation (107 citations)
- Extracting Class Activation Maps From Non-Discriminative Features As Well (88 citations)
- Weakly Supervised Semantic Segmentation via Adversarial Learning of Classifier and Reconstructor (88 citations)
- Semi-DETR: Semi-Supervised Object Detection With Detection Transformers (78 citations)
- Out-of-Candidate Rectification for Weakly Supervised Semantic Segmentation (74 citations)
- SOOD: Towards Semi-Supervised Oriented Object Detection (68 citations)
- Ambiguity-Resistant Semi-Supervised Learning for Dense Object Detection (66 citations)
and 22 more in the interactive map.
Tracking · 17 papers
- Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking (1,003 citations)
- SeqTrack: Sequence to Sequence Learning for Visual Object Tracking (462 citations)
- MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors (255 citations)
- MotionTrack: Learning Robust Short-Term and Long-Term Motions for Multi-Object Tracking (168 citations)
- Simple Cues Lead to a Strong Multi-Object Tracker (116 citations)
- Referring Multi-Object Tracking (104 citations)
- Unifying Short and Long-Term Tracking With Graph Hierarchies (97 citations)
- VideoTrack: Learning To Track Objects via Video Transformer (97 citations)
- Focus on Details: Online Multi-Object Tracking With Diverse Fine-Grained Representation (91 citations)
- UTM: A Unified Multiple Object Tracking Model With Identity-Aware Feature Enhancement (88 citations)
- Representation Learning for Visual Object Tracking by Masked Appearance Transfer (78 citations)
- OVTrack: Open-Vocabulary Multiple Object Tracking (69 citations)
and 5 more in the interactive map.