As it is cumbersome and expensive to acquire a huge amount of data for training neural dialog models, augmentation proposed effectively utilize existing samples. However, current techniques on the generation task mostly augment all cases in dataset without considering intrinsic attributes between different cases. We argue that not are beneficial task, suitable should obey following two attribut...

Pixel space augmentation has grown in popularity many Deep Learning areas, due to its effectiveness, simplicity, and low computational cost. Data for videos, however, still remains an under-explored research topic, as most works have been treating inputs stacks of static images rather than temporally linked series data. Recently, it shown that involving the time dimension when designing augment...

AbstractData Augmentation (DA) — generating extra training samples beyond the original set has been widely-used in today’s unbiased VQA models to mitigate language biases. Current mainstream DA strategies are synthetic-based methods, which synthesize new by either editing some visual regions/words, or re-generating them from scratch. However, these synthetic always unnatural and error-prone. To...

Determining the number of clusters is crucial for successful application clustering. In this paper, we propose a new order-determination method called data augmentation estimator (DAE), general model-based The based on novel idea that augments with an independently generated small cluster, which enables us to justify how instability clustering changes assumed in pattern provides alternative cha...

In this paper, we propose to apply data augmentation approaches that provide more diverse training images, thus helping train robust deep models for the Scene Text Recognition (STR) task. The methods are Random Blur Region (RBR) and Units (RBUs). Specifically, first introduce RBR designed STR training, randomly selects a region sets pixels in with an average value. However, when provides variou...