190 likes | 271 Views
Claim. Problem : SVM training is expensive Mining for hard negatives, bootstrapping Solution : LDA (Linear Discriminant Analysis). Extremely fast training, very similar performance. Linear Discriminant Analysis ( LDA) . Assumptions. Learning - Classification. Implementation.
E N D
Claim Problem: SVM training is expensive • Mining for hard negatives, bootstrapping Solution: LDA (Linear Discriminant Analysis). • Extremely fast training, very similar performance
Linear Discriminant Analysis (LDA) Assumptions Learning - Classification
Implementation Features a simple procedure that allows us to learn a and a (corresponding to the background) once, and then reuse it for every window size N and for every object category.
Implementation Mean Covariance
Regularization • Very large • In my experiments 10, for making sure that is PSD.
Clustering in WHO Space WHO HOG
Clustering in WHO Space WHO HOG
Pedestrian DetectionLinear Discriminant Models SVM LDA Cen
Summary • Whitened for HOG is better than HOG • LDA for fast training of hog templates • Object Independent Background (?) • mean better represents the cluster compared to the medoid • Use all the samples rather than 1 • Their statistical models also suggest that natural image statistics, largely ignored in the field of object detection, are worth (re)visiting.