Referring Expression Segmentation
The task aims at labeling the pixels of an image or video that represent an object instance referred by a linguistic expression. In particular, the referring expression (RE) must allow the identification of an individual object in a discourse or scene (the referent). REs unambiguously identify the target instance.
Papers
Showing 1–10 of 145 papers
All datasetsRefCoCo valRefCOCO testARefer-YouTube-VOS (2021 public validation)RefCOCO+ test BA2D SentencesRefCOCOg-valJ-HMDBDAVIS 2017 (val)RefCOCOg-testRefCOCO testBPhraseCutRefCOCO
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | HyperSeg | Overall IoU | 83.4 | — | Unverified |
| 2 | DeRIS-L | Overall IoU | 82.87 | — | Unverified |
| 3 | MLCD-Seg-7B | Overall IoU | 81.5 | — | Unverified |
| 4 | EVF-SAM | Overall IoU | 80.2 | — | Unverified |
| 5 | DETRIS | Overall IoU | 79 | — | Unverified |
| 6 | C3VG | Overall IoU | 77.86 | — | Unverified |
| 7 | VATEX | mIoU | 75.64 | — | Unverified |
| 8 | MaskRIS (Swin-B, combined DB) | Overall IoU | 75.1 | — | Unverified |
| 9 | MaskRIS (Swin-B) | Overall IoU | 73.96 | — | Unverified |
| 10 | EVP | Overall IoU | 72.94 | — | Unverified |