Skip to content

fix: the "table_index" is incorrect in LayoutParsingPipelineV2 #4678

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 3 commits into
base: release/3.3
Choose a base branch
from
Open

fix: the "table_index" is incorrect in LayoutParsingPipelineV2 #4678

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
1c0870d
fix: the table index is obtained by calculating the overlap ratio bet…
alanhsu777 Oct 28, 2025
3f934a9
fix: the table index is obtained by calculating the overlap ratio bet…
alanhsu777 Nov 4, 2025
2f7be6d
Merge branch 'PaddlePaddle:release/3.3' into fix/table-index-error
alanhsu777 Nov 4, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 9 additions & 3 deletions paddlex/inference/pipelines/layout_parsing/pipeline_v2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -689,11 +689,15 @@ def get_layout_parsing_objects(
- "block_bbox": The coordinates of the layout box.
"""

table_index = 0
table_cell_bboxes_list = [
[[int(pos) for pos in box] for box in table_res["cell_box_list"]]
for table_res in table_res_list
] if table_res_list else []
table_bboxes = [calculate_minimum_enclosing_bbox(table_cell_bboxes)
for table_cell_bboxes in table_cell_bboxes_list]
seal_index = 0
chart_index = 0
layout_parsing_blocks: List[LayoutBlock] = []

for box_idx, box_info in enumerate(layout_det_res["boxes"]):

label = box_info["label"]
Expand All @@ -703,8 +707,10 @@ def get_layout_parsing_objects(
block = LayoutBlock(label=label, bbox=block_bbox)

if label == "table" and len(table_res_list) > 0:
# The table index is obtained by calculating the overlap ratio between the bounding boxes of the tables.
table_overlap_ratio_list = [calculate_overlap_ratio(bbox1=block_bbox, bbox2=table_bbox) for table_bbox in table_bboxes]
table_index, table_overlap_ratio = max(enumerate(table_overlap_ratio_list), key=lambda x: x[1])
block.content = table_res_list[table_index]["pred_html"]
table_index += 1
elif label == "seal" and len(seal_res_list) > 0:
block.content = "\n".join(seal_res_list[seal_index]["rec_texts"])
seal_index += 1
Expand Down