Chinese Text Recognition with A Pre-Trained CLIP-Like Model Through Image-IDS Aligning
Open paper
Benchmark
Model
Rank
Results
scene-text-recognition-on-ctr-chinese-only
CCR-CLIP (extra data)
#5
Scene LACC: 71.31
Web LACC: 69.21
Document LACC: 98.29
…
Rank counts only results with a code link.