use rapidocr

machine-learning/immich_ml/schemas.py

@@ -25,6 +25,7 @@ class ModelTask(StrEnum):
     SEARCH = "clip"
     OCR = "ocr"
 
+
 class ModelType(StrEnum):
     DETECTION = "detection"
     RECOGNITION = "recognition"
@@ -32,6 +33,7 @@ class ModelType(StrEnum):
     VISUAL = "visual"
     OCR = "ocr"
 
+
 class ModelFormat(StrEnum):
     ARMNN = "armnn"
     ONNX = "onnx"
@@ -44,6 +46,7 @@ class ModelSource(StrEnum):
     OPENCLIP = "openclip"
     PADDLE = "paddle"
 
+
 ModelIdentity = tuple[ModelType, ModelTask]
 
 
@@ -87,19 +90,6 @@ class DetectedFace(TypedDict):
 FacialRecognitionOutput = list[DetectedFace]
 
 
-class OCROutput(TypedDict):
-    text: str
-    confidence: float
-    x1: int
-    y1: int
-    x2: int
-    y2: int
-    x3: int
-    y3: int
-    x4: int
-    y4: int
-
-
 class PipelineEntry(TypedDict):
     modelName: str
     options: dict[str, Any]