ruiliu-ai / FuseFormer

official Pytorch implementation of ICCV 2021 paper FuseFormer: Fusing Fine-Grained Information in Transformers for Video Inpainting.

ruiliu-ai/FuseFormer Issues

Question for the extra conv layers in Encoder
Updated 10 months ago
License
Updated a year ago
Appendix
Updated a year ago
License
Updated 2 years ago
about the detail of ref_ids
Updated 2 years ago
questions about save model
Updated 2 years ago2
RuntimeError: shape '[1, -1, 720, 512]' is invalid for input of size 28180480
Closed 2 years ago
Question about learning rate.
Updated 2 years ago
question about the random masks
Updated 2 years ago
How can I use multiple GPU to test
Updated 2 years ago
Input dimension limitation
Updated 2 years ago5
Bug of evaluate.py
Updated 2 years ago
Why if I set batchsize over 5 it will indicate me out of memory?
Closed 2 years ago1
Fine-tuning
Closed 2 years ago1
Are there video results of your work?
Closed 3 years ago1
Optical flow based warping error not implemented
Updated 2 years ago1
The trained model cannot be tested
Closed 3 years ago2
Evaluation Script
Closed 3 years ago1
Evaluation: Davis specification
Closed 3 years ago2
About your prepared Davis stationary masks sample numbers
Closed 3 years ago4
Whether there is a pretrained weight for discriminator?
Closed 3 years ago1
Error when running evaluate. py
Closed 3 years ago7
Big input resolution
Closed 3 years ago2
about the stationary mask for youtube-vos and davis test dataset
Closed 3 years ago4