administrator/frigate
1
0
Fork 0
mirror of https://github.com/blakeblackshear/frigate.git synced 2026-02-05 02:35:22 +03:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Add support for filtering labels for each detector in MetaDetectorConfig

Browse Source
This commit is contained in:
Sergey Krashevich 2023-04-26 03:33:25 +03:00
parent 5546c085fb
commit 55997dbbd5
No known key found for this signature in database
GPG Key ID: 625171324E7D3856
1 changed files with 29 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
frigate/detectors/plugins/meta.py
View File

@ -6,6 +6,7 @@ from frigate.detectors.detector_config import BaseDetectorConfig, ModelConfig
from frigate.util import deep_merge
from typing import List, Tuple, Dict, Any, Literal
from typing import Union
from typing import Optional
from typing_extensions import Annotated
from enum import Enum
from pydantic import Field, parse_obj_as
@ -28,6 +29,10 @@ class MetaDetectorConfig(BaseDetectorConfig):
default={"cpu": {"type": "cpu"}},
title="Detector hardware configuration.",
)
filtered_labels: Dict[str, Optional[List[str]]] = Field(
default={},
title="Labels to filter for each detector.",
)
max_detections: int = Field(
default=20,
title="Maximum number of detections to return after merging results",
@ -39,6 +44,8 @@ class MetaDetector(DetectionApi):
def __init__(self, meta_detector_config: MetaDetectorConfig):
self.max_detections = meta_detector_config.max_detections
self.filtered_labels = meta_detector_config.filtered_labels
self.labels = meta_detector_config.model.merged_labelmap
self.detectors = []
@ -62,12 +69,13 @@ class MetaDetector(DetectionApi):
"Customizing more than a detector model path is unsupported."
)
merged_model = deep_merge(
detector_config.model.dict(exclude_unset=True),
meta_detector_config.model.dict(exclude_unset=True),
detector_config.model.dict(exclude_unset=True),
)
detector_config.model = ModelConfig.parse_obj(merged_model)
meta_detector_config.detectors[key] = detector_config
self.detectors.append(self.create_detector(detector_config))
self.meta_detector_config = meta_detector_config
def merge_detections(self, detections_list: List[np.ndarray]) -> np.ndarray:
all_detections = np.vstack(detections_list)
@ -75,11 +83,28 @@ class MetaDetector(DetectionApi):
return all_detections[sorted_indices[: self.max_detections]]
def detect_raw(self, tensor_input) -> np.ndarray:
detections_list = [
detector.detect_raw(tensor_input) for detector in self.detectors
detections_list = []
for i, detector in enumerate(self.detectors):
detector_key = list(self.meta_detector_config.detectors.keys())[i]
filtered_labels = self.filtered_labels.get(detector_key)
detections = detector.detect_raw(tensor_input)
if filtered_labels is not None:
detections = np.array(
[
d
for d in detections
if self.get_label_name(d[0]) in filtered_labels
]
)
detections_list.append(detections)
return self.merge_detections(detections_list)
def get_label_name(self, index: int) -> str:
return self.labels.get(index)
def create_detector(self, detector_config):
current_module_name = os.path.splitext(os.path.basename(__file__))[0]
modules_folder = os.path.dirname(os.path.abspath(__file__))