VISAL-A novel learning strategy to address class imbalance.

In the imbalance data scenarios, Deep Neural Networks (DNNs) fail to generalize well on minority classes. In this letter, we propose a simple and effective learning function i.e, Visually Interpretable Space Adjustment Learning (VISAL) to handle the imbalanced data classification task. VISAL's objective is to create more room for the generalization of minority class samples by bringing in both the angular and euclidean margins into the cross-entropy learning strategy. When evaluated on the imbalanced versions of CIFAR, Tiny ImageNet, COVIDx and IMDB reviews datasets, our proposed method outperforms the state of the art works by a significant margin.

Depth camera based dataset of hand gestures.

The dataset contains RGB and depth version video frames of various hand movements captured with the Intel RealSense Depth Camera D435. The camera has two channels for collecting both RGB and depth frames at the same time. A large dataset is created for accurate classification of hand gestures under complex backgrounds. The dataset is made up of 29718 frames from RGB and depth versions corresponding to various hand gestures from different people collected at different time instances with complex backgrounds. Hand movements corresponding to scroll-right, scroll-left, scroll-up, scroll-down, zoom-in, and zoom-out are included in the data. Each sequence has data of 40 frames, and there is a total of 662 sequences corresponding to each gesture in the dataset. To capture all the variations in the dataset, the hand is oriented in various ways while capturing.