CLIP4STR-H (DFN-5B)

Reported on 6 benchmarks across 3 tasks · 1 paper · 6 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

Scene ParsingonSVT
Accuracy· uses extra data· 2023-05-23
99.1
SOTA
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model arXiv:2305.14014
Scene ParsingonWOST
1:1 Accuracy· 2023-05-23
90.9
SOTA
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model arXiv:2305.14014
Scene Text RecognitiononSVT
Accuracy· uses extra data· 2023-05-23
99.1
SOTA
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model arXiv:2305.14014
Scene Text RecognitiononWOST
1:1 Accuracy· 2023-05-23
90.9
SOTA
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model arXiv:2305.14014

Audio2 results

2D Semantic SegmentationonSVT
Accuracy· uses extra data· 2023-05-23
99.1
SOTA
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model arXiv:2305.14014
2D Semantic SegmentationonWOST
1:1 Accuracy· 2023-05-23
90.9
SOTA
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model arXiv:2305.14014