Alternating Back-Propagation for Generator Network

Tian Han* , Yang Lu* , Song-Chun Zhu , and Ying Nian Wu
* Equal contributions.

University of California, Los Angeles (UCLA), USA

Main > Incomplete Face

Recovery error

We design 5 experiments, with two types of occlusions: (1) 3 experiments are about salt and pepper occlusion, where we randomly place 3 × 3 masks on the 64 × 64 image domain to cover 50%, 70% and 90% of pixels respectively. These 3 experiments are denoted P.5, P.7, and P.9 respectively (P for pepper). (2) 2 experiments are about single region mask occlusion, where we place a 20 × 20 or 30 × 30 mask on the 64 × 64 image domain.

experiment P.5 P.7 P.9 M20 M30
error .0571 .0662 .0771 .0773 .1035

P.5 Results

Row 1: original images, not observed in learning. Row 2: training images. Row 3: recovered images during learning.

P.7 Results

Row 1: original images, not observed in learning. Row 2: training images. Row 3: recovered images during learning.

P.9 Results

Row 1: original images, not observed in learning. Row 2: training images. Row 3: recovered images during learning.

M20 Results

Row 1: original images, not observed in learning. Row 2: training images. Row 3: recovered images during learning.

M30 Results

Row 1: original images, not observed in learning. Row 2: training images. Row 3: recovered images during learning.

Back