scene/cameras.py

#
# Copyright (C) 2023, Inria
# GRAPHDECO research group, https://team.inria.fr/graphdeco
# All rights reserved.
#
# This software is free for non-commercial, research and evaluation use 
# under the terms of the LICENSE.md file.
#
# For inquiries contact  [email protected]
#

import torch
from torch import nn
import numpy as np
from utils.graphics_utils import getWorld2View2, getProjectionMatrix

class Camera(nn.Module):
    def __init__(self, colmap_id, intr, R, T, original_pose, FoVx, FoVy, image, gt_alpha_mask, dynamic_mask, enlarged_dynamic_mask,
                 dyna_avg_map, dyna_max_map, gt_dynamic_mask,
                 conf_map, depth_map,
                 image_name, uid,
                 trans=np.array([0.0, 0.0, 0.0]), scale=1.0, data_device = "cuda"
                 ):
        super(Camera, self).__init__()

        self.uid = uid
        self.colmap_id = colmap_id
        self.R = R
        self.T = T
        self.FoVx = FoVx
        self.FoVy = FoVy
        self.image_name = image_name

        try:
            self.data_device = torch.device(data_device)
        except Exception as e:
            print(e)
            print(f"[Warning] Custom device {data_device} failed, fallback to default cuda device" )
            self.data_device = torch.device("cuda")

        self.original_image = image.clamp(0.0, 1.0).to(self.data_device)
        self.image_width = self.original_image.shape[2]
        self.image_height = self.original_image.shape[1]
        
        if original_pose is not None:
            self.original_pose = torch.tensor(original_pose, dtype=torch.float32).to(self.data_device)

        if intr is not None:
            self.intr = intr
            
        if conf_map is not None:
            self.conf_map = conf_map.to(self.data_device)
        
        if depth_map is not None:
            self.depth_map = depth_map.to(self.data_device)
            
        if dynamic_mask is not None:
            self.dynamic_mask = dynamic_mask.to(self.data_device)
        
        if gt_dynamic_mask is not None:
            gt_dynamic_mask = gt_dynamic_mask.to(self.data_device)
            gt_dynamic_mask = gt_dynamic_mask.unsqueeze(0).repeat(3, 1, 1).unsqueeze(0).float()
            self.gt_dynamic_mask = torch.nn.functional.interpolate(
                    gt_dynamic_mask,
                    size=(self.image_height, self.image_width),
                    mode="nearest",
                ).squeeze(0)
        

        if enlarged_dynamic_mask is not None:
            self.enlarged_dynamic_mask = enlarged_dynamic_mask.to(self.data_device)

        if dyna_avg_map is not None:
            self.dyna_avg_map = dyna_avg_map.to(self.data_device)

        if dyna_max_map is not None:
            self.dyna_max_map = dyna_max_map.to(self.data_device)
            
        if gt_alpha_mask is not None:
            self.original_image *= gt_alpha_mask.to(self.data_device)
        else:
            self.original_image *= torch.ones((1, self.image_height, self.image_width), device=self.data_device)

        self.zfar = 100.0
        self.znear = 0.01

        self.trans = trans
        self.scale = scale
        
        self.world_view_transform = torch.tensor(getWorld2View2(R, T, trans, scale)).transpose(0, 1).cuda()
        self.projection_matrix = getProjectionMatrix(znear=self.znear, zfar=self.zfar, fovX=self.FoVx, fovY=self.FoVy).transpose(0,1).cuda()
        self.full_proj_transform = (self.world_view_transform.unsqueeze(0).bmm(self.projection_matrix.unsqueeze(0))).squeeze(0)
        self.camera_center = self.world_view_transform.inverse()[3, :3]

    def get_full_proj_transform(self, FoVx, FoVy):
        self.projection_matrix = getProjectionMatrix(znear=self.znear, zfar=self.zfar, fovX=FoVx, fovY=FoVy).transpose(0,1).cuda()
        return (self.world_view_transform.unsqueeze(0).bmm(self.projection_matrix.unsqueeze(0))).squeeze(0)

    def get_projection_matrix(self, FoVx, FoVy):
        return getProjectionMatrix(znear=self.znear, zfar=self.zfar, fovX=FoVx, fovY=FoVy).transpose(0,1).cuda()
class MiniCam:
    def __init__(self, width, height, fovy, fovx, znear, zfar, world_view_transform, full_proj_transform):
        self.image_width = width
        self.image_height = height    
        self.FoVy = fovy
        self.FoVx = fovx
        self.znear = znear
        self.zfar = zfar
        self.world_view_transform = world_view_transform
        self.full_proj_transform = full_proj_transform
        view_inv = torch.inverse(self.world_view_transform)
        self.camera_center = view_inv[3][:3]