« Erreur de reconstruction » : différence entre les versions


(Page créée avec « == en construction == == Définition == XXXXXXX Voir aussi '''autoencodeur''' == Français == ''' XXXXXX''' == Anglais == ''' Reconstruction Error''' ''Reconstruction error (in computer science) refers to the pixel level reconstruction error at a specific location in a frame of a video, which is calculated using a trained model. It represents the difference between the actual intensity level and the predicted intensity level by the model. Higher values o... »)
 
Aucun résumé des modifications
Ligne 12 : Ligne 12 :
''' Reconstruction Error'''
''' Reconstruction Error'''


''Reconstruction error (in computer science) refers to the pixel level reconstruction error at a specific location in a frame of a video, which is calculated using a trained model. It represents the difference between the actual intensity level and the predicted intensity level by the model. Higher values of the reconstruction error indicate a higher probability of detecting a video anomaly.''
''Reconstruction error (in computer science) refers to the pixel level reconstruction error at a specific location in a frame of a video, which is calculated using a trained model.''
 
''It represents the difference between the actual intensity level and the predicted intensity level by the model. Higher values of the reconstruction error indicate a higher probability of detecting a video anomaly.''


== Source ==
== Source ==

Version du 10 novembre 2024 à 16:16

en construction

Définition

XXXXXXX

Voir aussi autoencodeur

Français

XXXXXX

Anglais

Reconstruction Error

Reconstruction error (in computer science) refers to the pixel level reconstruction error at a specific location in a frame of a video, which is calculated using a trained model.

It represents the difference between the actual intensity level and the predicted intensity level by the model. Higher values of the reconstruction error indicate a higher probability of detecting a video anomaly.

Source

Source : Medium

Source : ScienceDirect

Source : ScienceDirect

Contributeurs: Arianne