Neural network-based human-computer interaction feature point annotation system and using method thereof

Abstract

The invention relates to the field of neural network data annotation, in particular to a neural network-based human-computer interaction feature point annotation system and a use method thereof, and the system comprises an input unit, a receiving unit, an annotation selection unit, an annotation withdrawing unit, an annotation moving unit and an annotation calibration unit; the input unit receivesan instruction input by a user, and receives a user instruction through the receiving unit to annotate a picture; a user uses the annotation selection unit for annotation selection, meanwhile, the annotation withdrawing unit is used for withdrawing annotations needing to be withdrawn, in the annotation process, the annotation moving unit is used for moving the annotations, and the annotations arecalibrated according to real values of data. Through the design of a computer screen, data is annotated, and the system is mainly used for annotating pictures of human body postures and has a very good annotation effect.

Description

technical field [0001] The invention relates to the field of human-computer labeling, in particular to a neural network-based human-computer interaction feature point labeling system and a method for using the same. Background technique [0002] In many neural network models, it is necessary to label the data well, but the current labeling equipment rarely has a machine that can specifically process the data, and the computer screen is also small, and it is also necessary to display the operation content when labeling the data. Corresponding increase, so the computer screen of dual display occurs, so that the space of the operable platform of computer increases. [0003] The existing dual-screen display computer screen is usually fixed on the bracket, and it is not convenient to fold and store when not in use, so it takes up space, and because the static electricity generated by the computer screen when in use will absorb dust and adhere to the computer screen, on the one ha...

AI Technical Summary

Problems solved by technology

[0002] In many neural network models, it is necessary to label the data well, but the current labeling equipment rarely has a machine that can specifically process the data, and the computer screen is also small, and it is also necessary to display the operation content when labeling the data. Corresponding increase, so dual-display computer screens appear, in order to increase the space of the computer's operable platform

[0003] The existing dual-screen display computer screen is usually fixed on the bracket, and it is not convenient to fold and store when not in use, so it takes up space, and because the static electricity generated by the computer screen when in use will absorb dust and adhere to the computer screen, on the one hand On the other hand, it is not easy to dissipate heat and clean the screen. In view of this, we propose a neural network-based human-computer interaction feature point labeling system and its application method

Images