◆ match_threshold
| float ailiaTracker.AiliaTracker.AILIATrackerSettings.match_threshold |
IoU threshold used for matching objects between frames.
Detections with an IoU greater than or equal to this value are treated as the same object.
Higher values make the matching stricter.
Default: 0.8
◆ nms_threshold
| float ailiaTracker.AiliaTracker.AILIATrackerSettings.nms_threshold |
Non‑Maximum Suppression (NMS) threshold used to remove duplicate detections.
Detections whose Intersection over Union (IoU) exceeds this value are merged.
Default: 0.7
◆ score_threshold
| float ailiaTracker.AiliaTracker.AILIATrackerSettings.score_threshold |
Minimum confidence score to accept a detection result.
Detections with scores below this value are ignored.
Default: 0.1
◆ track_buffer
| int ailiaTracker.AiliaTracker.AILIATrackerSettings.track_buffer |
Maximum number of frames to keep lost tracks in memory.
If the same object is detected again within this period, tracking is resumed.
Default: 30
◆ track_threshold
| float ailiaTracker.AiliaTracker.AILIATrackerSettings.track_threshold |
Confidence threshold for updating active tracks.
Normally, only detections with scores above this value are used to update existing tracks.
However, in the ByteTrack approach, detections with lower scores (between score_threshold and this value)
may also be used for updates if they achieve sufficient IoU matching with existing tracks.
A higher value makes the tracking stricter, while a lower value allows objects with temporarily low confidence to be maintained.
Default: 0.5
The documentation for this class was generated from the following file:
