Masked Transformer for Image Anomaly Localization.
Int J Neural Syst
; 32(7): 2250030, 2022 Jul.
Article
em En
| MEDLINE
| ID: mdl-35730477
ABSTRACT
Image anomaly detection consists in detecting images or image portions that are visually different from the majority of the samples in a dataset. The task is of practical importance for various real-life applications like biomedical image analysis, visual inspection in industrial production, banking, traffic management, etc. Most of the current deep learning approaches rely on image reconstruction the input image is projected in some latent space and then reconstructed, assuming that the network (mostly trained on normal data) will not be able to reconstruct the anomalous portions. However, this assumption does not always hold. We thus propose a new model based on the Vision Transformer architecture with patch masking the input image is split in several patches, and each patch is reconstructed only from the surrounding data, thus ignoring the potentially anomalous information contained in the patch itself. We then show that multi-resolution patches and their collective embeddings provide a large improvement in the model's performance compared to the exclusive use of the traditional square patches. The proposed model has been tested on popular anomaly detection datasets such as MVTec and head CT and achieved good results when compared to other state-of-the-art approaches.
Palavras-chave
Texto completo:
1
Coleções:
01-internacional
Base de dados:
MEDLINE
Assunto principal:
Processamento de Imagem Assistida por Computador
/
Tomografia Computadorizada por Raios X
Idioma:
En
Revista:
Int J Neural Syst
Assunto da revista:
ENGENHARIA BIOMEDICA
/
INFORMATICA MEDICA
Ano de publicação:
2022
Tipo de documento:
Article
País de afiliação:
Itália