| CPC G06V 20/52 (2022.01) [G06T 7/215 (2017.01); G06T 7/73 (2017.01); G06T 2207/30241 (2013.01)] | 15 Claims |
|
1. A method for determining images plausible to have a false negative object detection in an image sequence acquired by an image sensor, the method comprising:
providing a group of historic trajectories, wherein each historic trajectory comprises a reference track that represents a plurality of historic tracks and comprises an object class of historic object detections that belong to the plurality of historic tracks,
performing, by a tracker, tracking for determining tracks in the image sequence,
performing, by an object detector, object detection for determining object detections in the image sequence, wherein the object detector includes a neural network,
for a determined track that does not match any determined object detection, comparing the determined track with reference tracks of historic trajectories in the group of historic trajectories for identifying a matching reference track,
upon identifying a matching reference track, defining images of the determined track as being plausible to have a false negative object detection for the object class of the historic trajectory comprising the matching reference track, wherein the defined images are used to train the neural network of object detector, and
upon not identifying a matching reference track, defining the determined track as a false positive track.
|