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
No papers found.
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Ours (CoFormer+) | Top-1 Verb | 58.88 | — | Unverified |
| 2 | ClipSitu | Top-1 Verb | 58.19 | — | Unverified |
| 3 | CoFormer | Top-1 Verb | 44.66 | — | Unverified |
| 4 | SituFormer | Top-1 Verb | 44.2 | — | Unverified |
| 5 | Kernel GraphNet | Top-1 Verb | 43.27 | — | Unverified |
| 6 | GSRTR | Top-1 Verb | 40.63 | — | Unverified |
| 7 | JSL | Top-1 Verb | 39.94 | — | Unverified |
| 8 | ISL | Top-1 Verb | 39.36 | — | Unverified |
| 9 | CAQ + RE-VGG | Top-1 Verb | 38.19 | — | Unverified |
| 10 | GraphNet | Top-1 Verb | 36.72 | — | Unverified |