Data annotation method and device based on three-dimensional simulation scene and storage medium

Abstract

The invention provides a data annotation method and device based on a three-dimensional simulation scene and a storage medium, and the method comprises the steps: obtaining the three-dimensional simulation scene, extracting an image frame, and carrying out the preprocessing of the image frame, and obtaining an original scene coloring result; obtaining an annotation type and a corresponding coloring strategy, obtaining a target attribute parameter of each pixel of the target in the image frame according to the annotation type, and obtaining a coloring result according to the coloring strategy, wherein the annotation type comprises an illumination attribute, and the coloring strategy corresponding to the illumination attribute comprises the following steps: segmenting the time, uniformly processing image frames in each time period, annotating an area in an illumination range, and superposing annotation data at pixels corresponding to the image frames belonging to the same time period to obtain an illumination coloring result; and selecting a corresponding coloring result according to requirements, and superposing the coloring result with the original scene coloring result to obtain an output image. Therefore, the image frame can be dyed by taking the pixel as the unit, and the marking of the illumination thermodynamic diagram of the image frame is realized.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a data labeling method, device and storage medium based on a three-dimensional simulation scene. Background technique [0002] With the development of virtual reality and three-dimensional rendering technology, virtual scenes can already be made to look like real ones. In addition to bringing immersive sensory experience to people, the 3D simulation scene can also be used as a source of training data for computer vision and other related algorithms. The advantages of using 3D simulation scenes instead of actual scene shooting are: it can greatly reduce the workload of manual data collection; it can realize automatic data labeling that cannot be done in real shooting scenes; it can be combined with simulation algorithms to generate data under special events. The existing training data sets are usually obtained by manually acquiring images to be detected and manua...

AI Technical Summary

Problems solved by technology

Existing training data sets are usually obtained by manually acquiring images to be detected and manually annotating them, which is labor-intensive and inefficient. The method of using 3D simulation scenes to generate annotation data has yet to be developed.

Images