Method and apparatus for three-dimensional scene reconstruction

Abstract

A method and an apparatus for three-dimensional scene reconstruction are provided in the embodiment of the invention. The method includes: through a cascading neural networks formed by cascading a plurality of neural networks in a hierarchical form, a histogram correspond to each pixel point of that current processing area is obtained at a lower preset resolution based on a plurality of depth images, the sub-processing area corresponding to the current processing area and the voxel reconstruction data learned by the current neural network are determined by combining the current neural network,then the sub-processing area is taken as the processing object for the next neural network, and the above process is repeated on the basis of improving the separation rate of the processing object; and according to the voxel reconstruction data learned from each level of neural network, 3D scene reconstruction can be realized. The method and the apparatus process the processing area step by stepfrom the whole to the part in a gradual manner, greatly simplify the learning task of the neural network, remarkably reduce the calculation cost, and have higher efficiency, accuracy and practicability at the same time.

Description

technical field [0001] Embodiments of the present invention relate to the field of communication technologies, and more specifically, to a method and device for reconstructing a three-dimensional scene. Background technique [0002] High-quality reconstruction of 3D scenes has been one of the main frontiers of research in computer vision and computer graphics for many years. At the same time, the gradual popularization of consumer-grade color-depth (RGB-D) sensors (such as Microsoft Kinect and Intel RealSense) enables more novice users to perform 3D scanning of the surrounding environment, and also increases the robustness of users to 3D reconstruction algorithms. sexual needs. Despite the recent progress in 3D environment reconstruction, it is still a very challenging problem to reconstruct a high-accuracy 3D scene model from low-quality data obtained during casual scanning using consumer-grade RGB-D sensors. . [0003] Since neural networks have achieved impressive resu...

AI Technical Summary

Problems solved by technology

[0004] At present, some academic papers disclose some 3D reconstruction methods based on neural network learning, but these methods either require additional and time-consuming steps before local calculation, or need to optimize the intermediate results of 3D voxel representation globally, so that It is difficult for these methods to generate higher resolution results

Images