ML-CapsNet meets VB-DI-D: A novel distortion-tolerant baseline for perturbed object recognition.
Zhongqi LinZengwei ZhengJingdun JiaWanlin GaoFeng HuangPublished in: Eng. Appl. Artif. Intell. (2023)
Keyphrases
- object recognition
- maximum likelihood
- computer vision
- image understanding
- natural images
- object detection
- d objects
- object classification
- visual basic
- object categories
- object categorization
- image representation
- visual recognition
- image features
- vector quantizer
- relative improvement
- template matching
- feature matching
- feature detection
- em algorithm
- distortion correction
- constrained search
- database
- distorted images
- database programming
- information retrieval