Spatially-Regularized Features for Vehicle Re-Identification: An Explanation of Where Deep Models Should Focus.
Kai LvShuo WangShuai HanYoufang LinPublished in: IEEE Trans. Intell. Transp. Syst. (2023)
Keyphrases
- prior knowledge
- predictive power
- feature space
- classification models
- feature extraction
- predictive features
- abstraction levels
- co occurrence
- data sets
- false positives
- keypoints
- unsupervised feature learning
- machine learning approaches
- extracted features
- learning models
- pedestrian detection
- path planning
- statistical models
- spatial information
- image features
- probabilistic model
- feature vectors
- multiscale