| .. |
|
layout
|
c8a17c5f98
fix: rename doclayout_yolo.py to doclayoutyolo.py and add visualization method for bounding box results
|
2 月之前 |
|
mfd
|
cf2b74b030
fix: rename doclayout_yolo.py to doclayoutyolo.py and add visualization method for bounding box results
|
2 月之前 |
|
mfr
|
2fcffcb0af
fix: refactor image handling to use numpy arrays instead of PIL images
|
2 月之前 |
|
ocr
|
b7e4ea0b49
Reduce recognition batch size from 8 to 6 for improved OCR performance
|
1 月之前 |
|
ori_cls
|
6d5d1cf26b
Refactor image rotation handling in batch_analyze.py and paddle_ori_cls.py for improved compatibility with torch versions
|
1 月之前 |
|
reading_order
|
cbba27b4f5
refactor: reorganize project structure and update import paths
|
5 月之前 |
|
table
|
3c8b6e6b6b
Merge pull request #3499 from jinghuan-Chen/fix/fill_blank_rec_crop_empty_image
|
1 月之前 |
|
vlm_vllm_model
|
0758de6d24
Update vllm version and increase default GPU memory utilization
|
4 周之前 |
|
__init__.py
|
cbba27b4f5
refactor: reorganize project structure and update import paths
|
5 月之前 |
myhloli
0758de6d24
Update vllm version and increase default GPU memory utilization