Our method takes as input a collection of images (100 in our experiments) with known cameras, and outputs the volumetric density and normals, materials (BRDFs), and far-field illumination (environment map) of the scene.
When evaluating a trained model, there is a mismatch error when in the update of the alphamask of the sampler due to hardcoding to 265. The proposed change reads the grid_size directly from the checkpoint thus fixing this error.
When evaluating a trained model, there is a mismatch error when in the update of the alphamask of the sampler due to hardcoding to 265. The proposed change reads the grid_size directly from the checkpoint thus fixing this error.