Frequency-based Motion Representation
for Video Generative Adversarial Networks
Supplementary Material
Sangeek Hyun, Jaihyun Lew, Jiwoo Chung, Euiyeon Kim and Jae-Pil Heo
[Paper]
[Code]
1. Toy Dataset
Removing scenario on toy dataset.
Your browser does not support playing this Video
Original videos
Your browser does not support playing this Video
Low frequency only videos
Your browser does not support playing this Video
High frequency only videos
2. Randomly Sample Videos
Randomly generated videos on UCF-101, FaceForensics, VoxCeleb2-mini and BAIR-robot datasets.
3. Video Diversity
Each row indicates
different motion
codes with
same content code
.
Each row indicates
same motion
code with
different content
codes.
Your browser does not support playing this Video
BAIR-robot
Your browser does not support playing this Video
BAIR-robot
Your browser does not support playing this Video
FaceForensics
Your browser does not support playing this Video
FaceForensics
4. Temporal Interpolation
Interpolated videos on UCF-101, FaceForensics, VoxCeleb2-mini and BAIR-robot datasets.
Video interpolation on
BAIR-robot
dataset
Your browser does not support playing this Video
Original videos (4x slow)
Your browser does not support playing this Video
Interpolated Video
Video interpolation on
FaceForensics
dataset
Your browser does not support playing this Video
Original videos (4x slow)
Your browser does not support playing this Video
Interpolated Video
Video interpolation on
UCF-101
dataset
Your browser does not support playing this Video
Original videos (4x slow)
Your browser does not support playing this Video
Interpolated Video
Video interpolation on
VoxCeleb2-mini
dataset
Your browser does not support playing this Video
Original videos (4x slow)
Your browser does not support playing this Video
Interpolated Video
5. Temporal Extrapolation
Extrapolated videos on UCF-101, FaceForensics, VoxCeleb2-mini and BAIR-robot datasets.
Extrapolated Videos (4x longer): the
red box
denotes extrapolated frames
Your browser does not support playing this Video
UCF-101
Your browser does not support playing this Video
Faceforensics
Your browser does not support playing this Video
VoxCeleb2-mini
Your browser does not support playing this Video
BAIR-robot
6. Frequency-based Motion Manipulation
Each column
shares motion codes of
low frequency components
and
each row
shares motion codes of
high frequency components
.
Manipulated videos on
BAIR-robot
dataset
Your browser does not support playing this Video
Your browser does not support playing this Video
Your browser does not support playing this Video
Manipulated videos on
VoxCeleb2-mini
dataset
Your browser does not support playing this Video
Your browser does not support playing this Video
Your browser does not support playing this Video
Manipulated videos on
FaceForensics
dataset
Your browser does not support playing this Video
Your browser does not support playing this Video
Your browser does not support playing this Video
7. Manipulation - Removing
Removing scenario on
BAIR-robot
dataset
Your browser does not support playing this Video
Original videos
Your browser does not support playing this Video
Low frequency only videos
Your browser does not support playing this Video
High frequency only videos
Removing scenario on
FaceForensics
dataset
Your browser does not support playing this Video
Original videos
Your browser does not support playing this Video
Low frequency only videos
Your browser does not support playing this Video
High frequency only videos
Removing scenario on
VoxCeleb2-mini
dataset
Your browser does not support playing this Video
Original videos
Your browser does not support playing this Video
Low frequency only videos
Your browser does not support playing this Video
High frequency only videos
8. Manipulation - Altering
Each column indicates
videos altered a specific frequency range of motions
with the same content code.
The
blue box
denotes the original video.
Left column is altered
low frequency components
of motions.
Right column is altered
high frequency components
of motions.
Your browser does not support playing this Video
BAIR-robot (low frequency altering)
Your browser does not support playing this Video
BAIR-robot (high frequency altering)
Your browser does not support playing this Video
FaceForensics (low frequency altering)
Your browser does not support playing this Video
FaceForensics (high frequency altering)
Your browser does not support playing this Video
VoxCeleb2-mini (low frequency altering)
Your browser does not support playing this Video
VoxCeleb2-mini (high frequency altering)
Acknowledgements
This template was originally made by
Phillip Isola
and
Richard Zhang
for a
colorful
ECCV project; the code can be found
here
.