4 datasets found

Eye blinking detection in the wild, using spatiotemporal Pyramidal CNN with Depth-Wise Separable Convolution

The VSPW dataset is a large-scale dataset for video scene parsing in the wild, containing 2,806 videos with 480x853 pixel frames and 124 semantic categories.

The TrackingNet dataset is a benchmark for visual tracking, containing 511 video sequences with varying difficulties.

Deepfakes-generated videos raise not only significant social issues, but also ethical and privacy concerns. A recent report shows that nearly 96% of deepfake videos are porn...