SOTAVerified

Grounded Situation Recognition

Grounded Situation Recognition aims to produce the structured image summary which describes the primary activity (verb), its relevant entities (nouns), and their bounding-box groundings.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 11–15 of 15 papers

TitleStatusHype
Seeing Beyond Classes: Zero-Shot Grounded Situation Recognition via Language Explainer—0
Situation Recognition: Visual Semantic Role Labeling for Image UnderstandingCode0
ClipSitu: Effectively Leveraging CLIP for Conditional Predictions in Situation RecognitionCode0
Commonly Uncommon: Semantic Sparsity in Situation RecognitionCode0
GSRFormer: Grounded Situation Recognition Transformer with Alternate Semantic Attention RefinementCode0
Show:102550
← PrevPage 2 of 2Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Ours (CoFormer+)Top-1 Verb58.88—Unverified
2ClipSituTop-1 Verb58.19—Unverified
3CoFormerTop-1 Verb44.66—Unverified
4SituFormerTop-1 Verb44.2—Unverified
5Kernel GraphNetTop-1 Verb43.27—Unverified
6GSRTRTop-1 Verb40.63—Unverified
7JSLTop-1 Verb39.94—Unverified
8ISLTop-1 Verb39.36—Unverified
9CAQ + RE-VGGTop-1 Verb38.19—Unverified
10GraphNetTop-1 Verb36.72—Unverified