HyperAI
AccueilActualitésArticles de recherche récentsTutorielsEnsembles de donnéesWikiSOTAModèles LLMClassement GPUÉvénements
Recherche
À propos
HyperAI

Grounded Situation Recognition On Swig

Métriques

Top-1 Verb
Top-1 Verb u0026 Value
Top-5 Verbs
Top-5 Verbs u0026 Value

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle
Top-1 Verb
Top-1 Verb u0026 Value
Top-5 Verbs
Top-5 Verbs u0026 Value
Paper TitleRepository
CAQ + RE-VGG38.1930.2365.0550.21Attention-Based Context Aware Reasoning for Situation Recognition
CRF + Aug34.1226.4562.5946.88Commonly Uncommon: Semantic Sparsity in Situation Recognition
SituFormer44.235.2471.2155.75Rethinking the Two-Stage Framework for Grounded Situation Recognition
CRF32.3424.6458.8842.76Situation Recognition: Visual Semantic Role Labeling for Image Understanding-
JSL39.9431.4467.651.88Grounded Situation Recognition
GSRTR40.6332.1569.8154.13Grounded Situation Recognition with Transformers
Kernel GraphNet43.2735.4168.7255.62Mixture-Kernel Graph Attention Network for Situation Recognition-
GraphNet36.7227.5261.9045.39Situation Recognition with Graph Neural Networks
CoFormer44.6635.9873.3157.76Collaborative Transformers for Grounded Situation Recognition
Ours (CoFormer+)58.8851.10--Dynamic Scene Understanding from Vision-Language Representations-
ISL39.3630.0965.5150.16Grounded Situation Recognition
ClipSitu58.1947.2385.6968.42ClipSitu: Effectively Leveraging CLIP for Conditional Predictions in Situation Recognition
RNN + Fusion35.927.4563.0846.88Recurrent Models for Situation Recognition-
0 of 13 row(s) selected.