The dataset consists of a collection of **post-interview videos**, each containing multiple instances of micro-gestures. These **micro-gestures** are subtle, short-duration movements that may occur at different points throughout a video.

The same type of micro-gesture can appear more than once, often with only slight variations. Because the gestures may be very brief and not immediately obvious, careful and attentive annotation is required to capture them accurately.

Each participant is provided with a set of such videos for annotation. The goal is to identify the gesture class and its corresponding start and end times in each video.

⚠️ **Note**: The videos are provided solely for use in this assignment-competition. Redistribution, sharing, or any external use of the dataset is strictly prohibited. 

Please find the download link the exercise 1 introduction in https://moodle.oulu.fi/