A processor of an operation supporting device acquires pre-operation data including position information of a feature portion of an inside of a body cavity of a subject, which is generated before an operation, and a procedure in the operation, which is planned before the operation, and acquires an endoscope image generated in the operation. The processor generates real-time data including real-time position information of the feature portion of the inside of the body cavity, based at least on the pre-operation data and the endoscope image, and generates an image to be displayed on the endoscope image in a superimposed manner, based on the real-time data. The processor recognizes a real-time scene based at least on the procedure planned before the operation and the real-time data.
