Codesota · Models · DINOv2 (ViT-g/14)7 results · 7 benchmarks
Model card
DINOv2 (ViT-g/14).
unknown
§ 02 · Benchmarks
Every benchmark DINOv2 (ViT-g/14) has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | Pascal VOC 2012 | Computer Vision · Object Detection | miou | 86.2% | #2 | — | source ↗ |
| 02 | Cityscapes | Computer Vision · Semantic Segmentation | miou | 81.0% | #3 | — | source ↗ |
| 03 | UCF-101 | Computer Vision · Video classification | accuracy | 91.2% | #3 | — | source ↗ |
| 04 | Kinetics-400 | Computer Vision · Video classification | accuracy | 78.4% | #5 | — | source ↗ |
| 05 | Something-Something V2 | Computer Vision · Video classification | accuracy | 38.3% | #5 | — | source ↗ |
| 06 | ImageNet-1K | Computer Vision · Image Classification | accuracy | 86.5% | #9 | — | source ↗ |
| 07 | ADE20K | Computer Vision · Semantic Segmentation | miou | 53.0% | #10 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where DINOv2 (ViT-g/14) actually performs.
§ 04 · Papers
1 paper with results for DINOv2 (ViT-g/14).
- 2023-04-14· 7 results
DINOv2: Learning Robust Visual Features without Supervision
§ 06 · Sources & freshness
Where these numbers come from.
pwc-dump
7
results
0 of 7 rows marked verified.