Codesota · Computer Vision · Object Detection · LVIS v1.0Tasks/Computer Vision/Object Detection
Object Detection · benchmark dataset · 2019 · EN

Large Vocabulary Instance Segmentation v1.0.

1,203 object categories with federated, long-tail distribution across 164K COCO images. Tests real-world detection with rare and fine-grained categories.

Paper ↗Submit a result ↵
§ 01 · Leaderboard

Best published scores.

16 results indexed across 3 metrics. Shaded row marks current SOTA; ties broken by submission date.

Primary
mask-ap · higher is better
All metrics
box-ap, mask-ap, mask-ap-rare
box-ap
4 rows
#ModelOrgSubmittedPaper / codebox-ap
01DINO-XOSSIDEA ResearchNov 2024DINO-X: A Unified Vision Model for Open-World Object Det…71.40
02APE-LargeOSSTsinghua / MEGVIIDec 2023APE: Aligning and Prompting Everything All at Once for U…70.30
03Co-DINO (ViT-L)OSSSensetime / Sense-XNov 2022DETRs with Collaborative Hybrid Assignments Training68
04ViTDet-H (MAE)OSSMeta AIMar 2022Exploring Plain Vision Transformer Backbones for Object …64
mask-ap· primary
9 rows
#ModelOrgSubmittedPaper / codemask-ap
01DINO-XOSSIDEA ResearchNov 2024DINO-X: A Unified Vision Model for Open-World Object Det…67
02APE-LargeOSSTsinghua / MEGVIIDec 2023APE: Aligning and Prompting Everything All at Once for U…66.40
03InternImage-HOSSShanghai AI LabNov 2022InternImage: Exploring Large-Scale Vision Foundation Mod…65.40
04EVA-02-LOSSBAAI / PKUMar 2023EVA-02: A Visual Representation Powerhouse for Dense Rec…62.10
05ViTDet-H (MAE)OSSMeta AIMar 2022Exploring Plain Vision Transformer Backbones for Object …59.50
06Mask2Former (Swin-L)OSSMeta AI / UIUCDec 2021Masked-attention Mask Transformer for Universal Image Se…56.10
07ViTDet-HOSSMeta AIMar 2026arxiv53.40
08EVA-02-L (LVIS)OSSBAAIMar 2026arxiv50.70
09Mask2Former (Swin-L) LVISOSSMeta AIMar 2026arxiv44.60
mask-ap-rare
3 rows
#ModelOrgSubmittedPaper / codemask-ap-rare
01APE-LargeOSSTsinghua / MEGVIIDec 2023APE: Aligning and Prompting Everything All at Once for U…65.40
02EVA-02-LOSSBAAI / PKUMar 2023EVA-02: A Visual Representation Powerhouse for Dense Rec…61
03Mask2Former (Swin-L)OSSMeta AI / UIUCDec 2021Masked-attention Mask Transformer for Universal Image Se…53.50
Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.
§ 03 · Progress

5 steps
of state of the art.

Each row below marks a model that broke the previous record on mask-ap. Intermediate submissions are kept in the leaderboard above; only SOTA-setting entries are re-listed here.

Higher scores win. Each subsequent entry improved upon the previous best.

SOTA line · mask-ap
  1. Dec 2, 2021Mask2Former (Swin-L)Meta AI / UIUC56.10
  2. Mar 30, 2022ViTDet-H (MAE)Meta AI59.50
  3. Nov 10, 2022InternImage-HShanghai AI Lab65.40
  4. Dec 4, 2023APE-LargeTsinghua / MEGVII66.40
  5. Nov 21, 2024DINO-XIDEA Research67
Fig 3 · SOTA-setting models only. 5 entries span Dec 2021 → Nov 2024.
§ 04 · Literature

7 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies