Click any tag below to further narrow down your results
Links
Frigate lets you train lightweight MobileNetV2 models to classify tracked objects into finer categories (like identifying specific pets or detecting safety gear). The models require consistent predictions across multiple frames before assigning labels, which prevents false positives but means you need to carefully curate training data to avoid overfitting.
- Classification requires a two-step verification: confidence score above threshold (default 0.8) plus 60% consensus across at least 3 attempts, preventing single-frame misclassifications from triggering assignments.
- Training data quality matters far more than quantity—using dozens of nearly identical images actually degrades performance through overfitting, so you should start small with visually distinct examples and add hard cases (low confidence predictions, unusual angles/lighting) over time.
- The `none` class needs the opposite training approach from named classes: it should include visual diversity (small crops, partial views, false positives) so the model can recognize when something doesn't match your known categories, otherwise borderline detections get forced into named classes by default.
This doc explains how to configure which objects Frigate detects and tracks in video feeds, including how to filter false positives and mask detection zones. By default it only tracks people, but you can add cars, animals, and dozens of other object types.
- Frigate supports 90+ object types from Google Coral's dataset, but only tracks "person" by default — you must explicitly enable tracking for other objects like cars or dogs
- Object filters let you reduce false positives by setting minimum/maximum bounding box size, aspect ratio, and confidence thresholds per object type at global or per-camera level
- Mask zones prevent detection in specific frame areas (like trees or shadows) and can be applied globally or to individual object types