def compute_inv_depths(self, image, random_seed):
"""Computes inverse depth maps from single images"""
# Randomly flip and estimate inverse depth maps
flip_lr = random_seed < self.flip_lr_prob if self.training else False
inv_depths = make_list(flip_model(self.depth_net, image, flip_lr))
# If upsampling depth maps
if self.upsample_depth_maps:
inv_depths = interpolate_scales(
inv_depths, mode='nearest', align_corners=None)
# Return inverse depth maps
return inv_depths
def compute_inv_depths_feedback(self, image, disp, random_seed):
"""Computes inverse depth maps from single images"""
#concat image & disparity
# Randomly flip and estimate inverse depth maps
flip_lr = random.random() < self.flip_lr_prob if self.training else False
if flip_lr:
image = torch.flip(image,[3])
image = torch.cat([image, disp] , 1)
output = self.depth_net2(image)
inv_depths = make_list(output)
# If upsampling depth maps
if self.upsample_depth_maps:
inv_depths = interpolate_scales(
inv_depths, mode='nearest', align_corners=None)
# Return inverse depth maps
return inv_depths