mask parameter

[mowangmodi]

Hi, Xingyu. May I ask what the following mask parameters do? I don't understand the impact of these parameters. Could you please explain the following to me?

Loss parameters

parser.add_argument('--maskrs_max', type=float, default=0.0000015,
                    help='regularize mask size')# 
parser.add_argument('--maskrs_min', type=float, default=0.0000015,
                    help='regularize mask size')# 
parser.add_argument('--maskrs_k', type=float, default=0.9,
                    help='regularize mask size')# 
parser.add_argument('--maskrd', type=float, default=0.001,
                    help='regularize mask digit')
parser.add_argument('--weightKL', type=float, default=0.0000000000000001,
                    help='regularize encA')
parser.add_argument('--weightRecA', type=float, default=0.0000000001,
                    help='Rec A')
parser.add_argument('--weightMS', type=float, default=0.0000000001,
                    help='mode seeking') 

The parameters you provided for the example: --maskrs_max 5e-2 --maskrs_min 6e-3 --maskrs_k 1e-3 --maskrd 0 --weightKL 1e-5

[rover-xingyu]

Hi, the parameters of maskrs_max, maskrs_min and maskrs_k are used for ExponentialAnnealingWeight, the maskrd is used to make the mask to be closer to 1/0, and the weightKL is used to regularize the distribution of embeddings.

[mowangmodi]

Why was the MLP weight of the mask not loaded during testing?

[rover-xingyu]

The mask is only used for masking out occlusions during training. For testing, you can directly render the 3D scene representation to get free-occlusion images.