Skip to content

ECCV2020-RAFT(Recurrent All Pairs Field Transforms for Optical Flow) tensorflow/tensorpack implementation

Notifications You must be signed in to change notification settings

gonglixue/RAFT-tf

Folders and files

NameName
Last commit message
Last commit date

Latest commit

858082a · Nov 24, 2020

History

7 Commits
Sep 18, 2020
Sep 18, 2020
Nov 24, 2020
Nov 24, 2020
Sep 18, 2020
Sep 18, 2020
Sep 18, 2020
Sep 18, 2020
Sep 18, 2020
Sep 18, 2020
Sep 18, 2020
Sep 18, 2020
Sep 18, 2020

Repository files navigation

RAFT-tensorflow/tensorpack

This repository contains the tensorflow-implementation(with tensorpack) for this paper:

RAFT:Recurrent All Pairs Field Transforms for Optifal Flow

Official Pytorch Implementation

TODO

  • Basic inference code.
  • grid_sample align_corners=True in tf-implementation.
  • Add cuda extension for efficent correlation calculation.
  • Check if the batch_size could be free.
  • Reproduce the training process.

Requirements

tensorflow-gpu >= 1.14
opencv-python
numpy
tensorpack
CUDA 10.1

Demo

  1. Download the pretrained model from GoogleDrive to release_weight folder. The *.npz files are converted from the official pytorch *.pth model provided in the official repository.

  2. Run the inference demo:

bash ./infer_image.sh

or

python infer_raft.py --im1 frame_0016.png --im2 frame_0017.png --load release_weight/raft-things.npz

python infer_raft.py --im1 frame_0016.png --im2 frame_0017.png --load release_weight/raft-small.npz --small

Note

The inference result of my tensorflow implementation is as below. There is still a few of differences from the official implementation. I will continue to follow up.

My tf-implementation Official pytorch-implementation
my offcial

About

ECCV2020-RAFT(Recurrent All Pairs Field Transforms for Optical Flow) tensorflow/tensorpack implementation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published