SOTAVerified

Weakly Supervised Object Detection

Weakly Supervised Object Detection (WSOD) is the task of training object detectors with only image tag supervisions.

( Image credit: Soft Proposal Networks for Weakly Supervised Object Localization )

Papers

Showing 125 of 142 papers

TitleStatusHype
WeakSAM: Segment Anything Meets Weakly-supervised Instance-level RecognitionCode2
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized CutCode2
Semi-Weakly Supervised Object Detection by Sampling Pseudo Ground-Truth BoxesCode1
Multiple instance learning on deep features for weakly supervised object detection with extreme domain shiftsCode1
PDL: Regularizing Multiple Instance Learning with Progressive Dropout LayersCode1
ImaginaryNet: Learning Object Detectors without Real Images and AnnotationsCode1
LEOD: Label-Efficient Object Detection for Event CamerasCode1
Active Learning Strategies for Weakly-supervised Object DetectionCode1
H2FA R-CNN: Holistic and Hierarchical Feature Alignment for Cross-Domain Weakly Supervised Object DetectionCode1
Multiple Instance Detection Network with Online Instance Classifier RefinementCode1
Object Instance Mining for Weakly Supervised Object DetectionCode1
PCL: Proposal Cluster Learning for Weakly Supervised Object DetectionCode1
Harmonizing Transferability and Discriminability for Adapting Object DetectorsCode1
Scaling Novel Object Detection with Weakly Supervised Detection TransformersCode1
Max Pooling with Vision Transformers reconciles class and shape in weakly supervised semantic segmentationCode1
Domain-Adaptive Object Detection via Uncertainty-Aware Distribution AlignmentCode1
Boosting Weakly Supervised Object Detection with Progressive Knowledge TransferCode1
Distilling Knowledge from Refinement in Multiple Instance Detection NetworksCode1
Comprehensive Attention Self-Distillation for Weakly-Supervised Object DetectionCode1
Boosting Weakly Supervised Object Detection via Learning Bounding Box AdjustersCode1
Domain-Adaptive Self-Supervised Pre-Training for Face & Body Detection in DrawingsCode1
Cross-Domain Weakly-Supervised Object Detection through Progressive Domain AdaptationCode1
Cyclic-Bootstrap Labeling for Weakly Supervised Object DetectionCode1
Instance-aware, Context-focused, and Memory-efficient Weakly Supervised Object DetectionCode1
Object Discovery via Contrastive Learning for Weakly Supervised Object DetectionCode1
Show:102550
← PrevPage 1 of 6Next →

No leaderboard results yet.