We are presenting VideoINR in
#CVPR2022
, a new continuous Video Implicit Neural Representation, that allows any scale super-resolution / interpolation in both space and time.
Paper:
Project page:
1/n)
3
19
139
Replies
We can zoom in the video frames in any scale, or replay/interpolate the video in any framerate, see the following video:
Original Input, Bilinear resize, VideoINR (ours)
2/n)
1
0
5
Or we can zoom in space and interpolate in time at the same time, see the following video:
Original Input, Bilinear resize, VideoINR (ours)
3/n)
1
0
5
This work is done mainly by our intern Zeyuan Chen (who is joining UCSD for his Ph.D. study in Zhuowen Tu's lab), and in collaboration with
@YinboChen
, and the
@PicsartAI
team led by
@humphrey_shi
4/n)
1
0
4
code is also released:
5/n)
0
0
2
