Weakly Supervised Object Detection On Pascal

評価指標

MAP

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名	MAP	Paper Title	Repository
MSLPD	35.4	Few-Example Object Detection with Model Communication
WCCN	37.9	Weakly Supervised Cascaded Convolutional Networks	-
wetectron(single-model)	52.1	Instance-aware, Context-focused, and Memory-efficient Weakly Supervised Object Detection
Pred Net (Ens)	49.5	Dissimilarity Coefficient based Weakly Supervised Object Detection	-
C-MIDN+FRCNN	50.3	C-MIDN: Coupled Multiple Instance Detection Network With Segmentation Guidance for Weakly Supervised Object Detection	-
WeakSAM-OICR-DINO (with SAM)	63.7	WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition
WeakSAM-MIST-DINO (with SAM)	70.2	WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition
NSOD	36.6	Training Object Detectors from Few Weakly-Labeled and Many Unlabeled Images	-
OICR + W-RPN	43.2	You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised Object Detection	-
OICR-Ens + FRCNN	42.5	Multiple Instance Detection Network with Online Instance Classifier Refinement
ZLDN-L	42.9	Zigzag Learning for Weakly Supervised Object Detection	-
CASD(VGG16)	53.6	Comprehensive Attention Self-Distillation for Weakly-Supervised Object Detection
WeakSAM-MIST-Faster RCNN (with SAM)	69.2	WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition
Deep Self-Taught Learning	38.3	Deep Self-Taught Learning for Weakly Supervised Object Localization	-
MELM	42.4	Min-Entropy Latent Model for Weakly Supervised Object Detection
OIM+IR+FRCNN	46.4	Object Instance Mining for Weakly Supervised Object Detection
WSDDN + context	35.3	ContextLocNet: Context-Aware Deep Network Models for Weakly Supervised Localization
WeakSAM-MIST (with SAM)	66.9	WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition
LM-VGPMIL	37.8	Variational Bayesian Multiple Instance Learning With Gaussian Processes	-
Ours+FRCNN	48.0	Utilizing the Instability in Weakly Supervised Object Detection	-

0 of 32 row(s) selected.