Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

LVIS dataset and benchmarks – Page 2 · SOTA2 Research

Research Datasets

LVIS

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Open-Vocabulary Object Detection

LVIS

42.1APr

26

Image Segmentation

LVIS LQ-1 (test)

64.3IoU

24

Image Segmentation

LVIS LQ-2 (test)

62.2IoU

24

Image Segmentation

LVIS LQ-3 (test)

58.4IoU

24

Instance Segmentation

OV-LVIS

24.8AP (Rare)

23

Referring Object Classification

LVIS (test)

73.39Accuracy

22

Open-vocabulary detection

LVIS OV (val)

50.1APr

20

Bounding-box classification

LVIS

38.3Top-1 Accuracy

19

Instance Segmentation

OV-LVIS v1.0 (val)

32.3mAP (Rare, Mask)

19

Bounding box classification

LVIS 1203

47.3Top-1 Accuracy

18

Category-level image recognition

LVIS

93.6Similarity Score

18

Referring Expression Grounding

LVIS-Ref

87.7Accuracy (Box)

18

Object Detection

LVIS (minival5k)

44.7AP

18

Object Detection

LVIS zero-shot (mini val)

31.9AP (Common)

17

In-Context Segmentation

LVIS-92i

59mIoU

16

Object Detection

LVIS full (val)

50.4AP all (old)

16

Object Detection

LVIS v1 (eval)

27.5AP

15

Object Detection

OV-LVIS (test)

34.9mAP_r

14

Object Counting

LVIS-Count 100-image zero-shot (test)

23.36MAE

13

Object Detection

OV-LVIS (val)

25.5AP (Novel)

13

Open-vocabulary Object Detection

LVIS MDETR split (minival)

55.8AP

13

Object Localisation

LVIS subset with <= 3 objects (test)

26mIoU

13

Few-Shot Segmentation

LVIS 92^i

51.9mIoU

12

Image-Text Retrieval

LVIS

22Recall@1

12

Object Detection

LVIS v1.0

29.6APbb

12

Page 2 of 6

Popular tasks

Object Detection Instance Segmentation Open-vocabulary object detection Pose Estimation Image Segmentation

Follow for updates

Join SOTA2 Contact