helpy4/immich
1
0
Fork 0
mirror of https://github.com/immich-app/immich synced 2025-10-17 18:19:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

refactor(ocr): update OCR schema and response structure to use individual coordinates instead of bounding box, and adjust related service and repository files

Browse source
This commit is contained in:
CoderKang 2025-06-02 22:26:49 +08:00 • committed by mertalev
parent 46ef02342d
commit df36a09cd3
No known key found for this signature in database
GPG key ID: DF6ABC77AAD98C95
7 changed files with 66 additions and 34 deletions
Download patch file Download diff file Expand all files Collapse all files

8
machine-learning/immich_ml/models/ocr/paddle.py
View file

@ -34,13 +34,17 @@ class PaddleOCRecognizer(InferenceModel):
valid_texts_and_scores = [
(text, score, box)
for result in results
for text, score, box in zip(result['rec_texts'], result['rec_scores'], result['rec_boxes'].tolist())
for text, score, box in zip(result['rec_texts'], result['rec_scores'], result['rec_polys'])
if score >= self.min_score
]
if not valid_texts_and_scores:
return []
return [
OCROutput(text=text, confidence=score, boundingBox={"x1": box[0], "y1": box[1], "x2": box[2], "y2": box[3]})
OCROutput(
text=text, confidence=score,
x1=box[0][0], y1=box[0][1], x2=box[1][0], y2=box[1][1],
x3=box[2][0], y3=box[2][1], x4=box[3][0], y4=box[3][1]
)
for text, score, box in valid_texts_and_scores
]