hitachinsk / FGT

[ECCV 2022] Flow-Guided Transformer for Video Inpainting

https://hitachinsk.github.io/publication/2022-10-01-Flow-Guided-Transformer-for-Video-Inpainting

Repository from Github https://github.comhitachinsk/FGT

hitachinsk/FGT Issues

Constantly getting 'ValueError: need at least one array to stack' error
Closed 2 years ago1
Slow speed when inferencing
Updated 7 months ago3
CUDA error
Updated 7 months ago1
FGT++
Updated 7 months ago5
Watermark removal results is worse
Updated 8 months ago
Which one is better？ FGT or ISVI？
Closed 8 months ago2
Outpainting Training on Custom Data
Updated 9 months ago
Can't be cleaned up, is it an issue with the pre-trained model?
Closed a year ago4
Not inpainting beginning and end of videos
Closed a year ago
Cuda OOM (1k+ frame video) on A100
Closed a year ago4
supplementary material
Closed 2 years ago2
Add a way to extract an `ONNX` file
Updated 2 years ago1
About square mask settings
Closed 2 years ago1
The flow propagation
Closed 2 years ago3
output video size
Closed 2 years ago3
Questions about the correspondence between the published pre-trained models and the algorithms in the paper.
Closed 2 years ago3
T dimension problems in encoders
Closed 2 years ago4
What is the GPU model used in the experiment?
Closed 2 years ago2
Run on image
Closed 2 years ago2
About FGT++
Closed 2 years ago2
How to fine-tune FGT on my own dataset
Closed 2 years ago2
python3.6.8 cannot install numpy=1.22.2,Can I upgrade python version or downgrade numpy version
Closed 2 years ago1
Default watermark removal mode config not work very well
Closed 2 years ago2
test result resolution
Closed 2 years ago1
RuntimeError:cuda out of memory
Closed 2 years ago2
About DDP aggregation gradients on different GPUs
Closed 2 years ago2
UserWarning of `scheduler.step()` and warning of Grad strides do not match bucket view strides
Closed 2 years ago3
About setting parameter batch_size=1, num_frames=5
Updated 2 years ago3
Organization of the YouTube-VOS dataset in the ‘myData’ folder for LAFC Net training?
Closed 2 years ago12
test own video , need mask?
Updated 2 years ago1
How can I test a custom video? I have only any video.
Closed 3 years ago3
FGT model config problem
Closed 3 years ago4
yaml.load fails in wsl, change load method to make it work
Closed 3 years ago2
Add python to env creation so correct pip will be used
Closed 3 years ago3
add models to Hugging Face Hub
Updated 3 years ago2
Is paper released?
Closed 3 years ago1
About image resolution and optical flow resolution
Closed 3 years ago2
Pre-training model download problem
Closed 3 years ago3