LVIS is a dataset for long tail instance segmentation. It has annotations for over 1000 object categories in 164k images.
Source: LVISBenchmarks
| Trend | Task | Dataset Variant | Best Model | Paper | Code |
|---|---|---|---|---|---|
|
LVIS v1.0
|
DITO
|
|||
|
LVIS v1.0 val
|
Co-DETR
|
|||
|
LVIS v1.0 val
|
Co-DETR
|
|||
|
LVIS v1.0 val
|
Organizer Provided Baseline
|
|||
|
LVIS v1.0 minival
|
OWLv2
|
|||
|
LVIS v1.0 val
|
Cooperative Foundational Models
|
|||
|
LVIS v1.0 test-dev
|
Asynchronous SSL
|
|||
|
LVIS v1.0 val
|
OWLv2
|
|||
|
LVIS v1.0 val
|
HQ-SAM
|
|||
|
LVIS v1.0 minival
|
Co-DETR
|
|||
|
LVIS v1.0 val
|
R101 Faster R-CNN
|
|||
|
LVIS v1.0
|
ScaleDet
|
|||
|
LVIS v1.0 val
|
CutLER
|
|||
|
LVIS v1.0 test-dev
|
R50-FPN-MaskRCNN-TTA
|
Papers
| Paper | Code | Results | Date | Stars |
|---|
