Model card
Davar-Lab-OCR.
Hikvision Research Institutetable-recognition
ICDAR 2021 Scientific Literature Parsing Task B winner (1st place). Hierarchical encoder-decoder for table structure recognition on PubTabNet.
§ 01 · Benchmarks
Every benchmark Davar-Lab-OCR has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | table-recognition-challenge-mini-test | Computer Vision · Table Recognition | teds-all-samples | 96.4% | #3 | — | source ↗ |
| 02 | table-recognition-challenge-mini-test | Computer Vision · Table Recognition | teds-complex-samples | 94.8% | #4 | — | source ↗ |
| 03 | table-recognition-challenge-mini-test | Computer Vision · Table Recognition | teds-simple-samples | 97.9% | #4 | — | 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.
§ 02 · Strengths by area
Where Davar-Lab-OCR actually performs.
§ 05 · Sources & freshness
Where these numbers come from.
paper
3
results
0 of 3 rows marked verified.