License plate recognition (ALPR) backend (#14564)

* Update version * Face recognition backend (#14495) * Add basic config and face recognition table * Reconfigure updates processing to handle face * Crop frame to face box * Implement face embedding calculation * Get matching face embeddings * Add support face recognition based on existing faces * Use arcface face embeddings instead of generic embeddings model * Add apis for managing faces * Implement face uploading API * Build out more APIs * Add min area config * Handle larger images * Add more debug logs * fix calculation * Reduce timeout * Small tweaks * Use webp images * Use facenet model * Improve face recognition (#14537) * Increase requirements for face to be set * Manage faces properly * Add basic docs * Simplify * Separate out face recognition frome semantic search * Update docs * Formatting * Fix access (#14540) * Face detection (#14544) * Add support for face detection * Add support for detecting faces during registration * Set body size to be larger * Undo * Update version * Face recognition backend (#14495) * Add basic config and face recognition table * Reconfigure updates processing to handle face * Crop frame to face box * Implement face embedding calculation * Get matching face embeddings * Add support face recognition based on existing faces * Use arcface face embeddings instead of generic embeddings model * Add apis for managing faces * Implement face uploading API * Build out more APIs * Add min area config * Handle larger images * Add more debug logs * fix calculation * Reduce timeout * Small tweaks * Use webp images * Use facenet model * Improve face recognition (#14537) * Increase requirements for face to be set * Manage faces properly * Add basic docs * Simplify * Separate out face recognition frome semantic search * Update docs * Formatting * Fix access (#14540) * Face detection (#14544) * Add support for face detection * Add support for detecting faces during registration * Set body size to be larger * Undo * initial foundation for alpr with paddleocr * initial foundation for alpr with paddleocr * initial foundation for alpr with paddleocr * config * config * lpr maintainer * clean up * clean up * fix processing * don't process for stationary cars * fix order * fixes * check for known plates * improved length and character by character confidence * model fixes and small tweaks * docs * placeholder for non frigate+ model lp detection --------- Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
2026-10-12 17:52:48 +03:00 · 2025-02-08 12:47:01 -06:00
parent 594a4e0ba3
commit c4ce7f9800
10 changed files with 1151 additions and 9 deletions
@@ -77,6 +77,10 @@ class Embeddings:
            if config.semantic_search.model_size == "large"
            else "jinaai/jina-clip-v1-vision_model_quantized.onnx",
            "jinaai/jina-clip-v1-preprocessor_config.json",
+            "facenet-facenet.onnx",
+            "paddleocr-onnx-detection.onnx",
+            "paddleocr-onnx-classification.onnx",
+            "paddleocr-onnx-recognition.onnx",
        ]

        for model in models:
@@ -138,6 +142,47 @@ class Embeddings:
                device="GPU",
            )

+        self.lpr_detection_model = None
+        self.lpr_classification_model = None
+        self.lpr_recognition_model = None
+
+        if self.config.lpr.enabled:
+            self.lpr_detection_model = GenericONNXEmbedding(
+                model_name="paddleocr-onnx",
+                model_file="detection.onnx",
+                download_urls={
+                    "detection.onnx": "https://github.com/hawkeye217/paddleocr-onnx/raw/refs/heads/master/models/detection.onnx"
+                },
+                model_size="large",
+                model_type=ModelTypeEnum.alpr_detect,
+                requestor=self.requestor,
+                device="CPU",
+            )
+
+            self.lpr_classification_model = GenericONNXEmbedding(
+                model_name="paddleocr-onnx",
+                model_file="classification.onnx",
+                download_urls={
+                    "classification.onnx": "https://github.com/hawkeye217/paddleocr-onnx/raw/refs/heads/master/models/classification.onnx"
+                },
+                model_size="large",
+                model_type=ModelTypeEnum.alpr_classify,
+                requestor=self.requestor,
+                device="CPU",
+            )
+
+            self.lpr_recognition_model = GenericONNXEmbedding(
+                model_name="paddleocr-onnx",
+                model_file="recognition.onnx",
+                download_urls={
+                    "recognition.onnx": "https://github.com/hawkeye217/paddleocr-onnx/raw/refs/heads/master/models/recognition.onnx"
+                },
+                model_size="large",
+                model_type=ModelTypeEnum.alpr_recognize,
+                requestor=self.requestor,
+                device="CPU",
+            )
+
    def embed_thumbnail(
        self, event_id: str, thumbnail: bytes, upsert: bool = True
    ) -> ndarray: