An image processing device includes a camera interface connectable to a camera configured to capture a video of a worker in a work area, an output interface connectable to a display device that is viewable from the work area, a memory that stores task information about a task to be performed by the worker, and a processor configured to determine a position and a posture of the worker in the work area based on the captured video, determine whether the worker has performed the task properly based on the task information stored in the memory and the determined position and posture of the worker, upon determining that the worker has not performed the task properly, extract from the video a first part showing the task that has been performed by the worker, and control the output interface to output the first part to the display device.