Method, system and storage medium for three-dimensional reconstruction

The invention relates to a method, a system and a storage medium for three-dimensional reconstruction. Various embodiments related to three-dimensional reconstruction are described. In one embodiment, the method for training the three-dimensional reconstruction model comprises the following steps: generating an initial voxel envelope of a target object based on an image obtained by shooting the target object at multiple view angles; randomly sampling points in the initial voxel envelope to obtain a set of sampling points; performing global feature extraction on the image to obtain a global feature map; determining global features corresponding to the sampling points from the global feature map based on geometric association; performing encoding processing on the geometric information about the sampling points to generate geometric encoding information; and training the model based at least on the global features and the geometric coding information. 本公开内容涉及用于三维重建的方法、系统和存储介质。描述了关于三维重建的各种实施例。在一个实施例中，用于训练三维重建模型的