UnusualAction Dataset

UnusualAction Dataset for Action Recognition Nitika Nigam, Tanima Dutta and Hari Prabhat Gupta Indian Institute of Technology (BHU), India. Overview: UnusualAction is an uncertain action recognition dataset that rarely happens and collected from YouTube. The dataset comprises 14 unusual action categories, and each category contains 50-100 videos. UnusualAction gives the diversity in terms of different falling actions and with the presence of noises, such as, variations in camera motions, person appearance, viewpoint, cluttered background, illumination conditions, etc. It is a challenging dataset for uncertain action recognition. Most action recognition datasets are based on certain actions; on the contrary, UnusualAction aims to encourage further research into uncertain action recognition by learning and exploring new realistic action categories. Structure for UnusualAction Dataset ● Data associated with each UnusualAction category is stored in separate directories. ● Each directory comprises *.mp4 or *.avi files for videos. ● The directory is arranged in the following structure: FallAction_datasets ├──Blending_phone ├── Crushing_laptop ├── Cutting_keyboard ├── Drilling_Laptop ├── Drilling_Phone ├── Frying_Phone ├── Hammering_Laptop ├── Hammering_phone ├── Hammering_pumpkin ├── Hammering_watermelon ├── Microwave_shoes ├──Microwave_phone ├── Washing_laptop ├── Washing_Paptop