Image Data Augmentation

Mixup

Introduced by Zhang et al. in mixup: Beyond Empirical Risk Minimization

Mixup is a data augmentation technique that that generates a weighted combinations of random image pairs from the training data. Given two images and their ground truth labels: $\left(x_{i}, y_{i}\right), \left(x_{j}, y_{j}\right)$, a synthetic training example $\left(\hat{x}, \hat{y}\right)$ is generated as:

$$ \hat{x} = \lambda{x_{i}} + \left(1 − \lambda\right){x_{j}} $$ $$ \hat{y} = \lambda{y_{i}} + \left(1 − \lambda\right){y_{j}} $$

where $\lambda \sim \text{Beta}\left(\alpha = 0.2\right)$ is independently sampled for each augmented example.

Source: mixup: Beyond Empirical Risk Minimization

Papers


Paper Code Results Date Stars

Components


Component Type
🤖 No Components Found You can add them if they exist; e.g. Mask R-CNN uses RoIAlign

Categories