All Projects → linchaobao → EPPM

linchaobao / EPPM

Licence: MIT license
CUDA implementation of the paper "Fast Edge-Preserving PatchMatch for Large Displacement Optical Flow" in CVPR 2014.

Programming Languages

Cuda
1817 projects
C++
36643 projects - #6 most used programming language
c
50402 projects - #5 most used programming language
CMake
9771 projects

Projects that are alternatives of or similar to EPPM

Joint-Motion-Estimation-and-Segmentation
[MICCAI'18] Joint Learning of Motion Estimation and Segmentation for Cardiac MR Image Sequences
Stars: ✭ 45 (+32.35%)
Mutual labels:  optical-flow
video features
Extract video features from raw videos using multiple GPUs. We support RAFT and PWC flow frames as well as S3D, I3D, R(2+1)D, VGGish, CLIP, ResNet features.
Stars: ✭ 225 (+561.76%)
Mutual labels:  optical-flow
correlation flow
ROS package for Correlation Flow (ICRA 2018)
Stars: ✭ 28 (-17.65%)
Mutual labels:  optical-flow
mmflow
OpenMMLab optical flow toolbox and benchmark
Stars: ✭ 711 (+1991.18%)
Mutual labels:  optical-flow
flow1d
[ICCV 2021 Oral] High-Resolution Optical Flow from 1D Attention and Correlation
Stars: ✭ 91 (+167.65%)
Mutual labels:  optical-flow
PyTrx
PyTrx is a Python object-oriented programme created for the purpose of calculating real-world measurements from oblique images and time-lapse image series. Its primary purpose is to obtain velocities, surface areas, and distances from oblique, optical imagery of glacial environments.
Stars: ✭ 31 (-8.82%)
Mutual labels:  optical-flow
Unflow
UnFlow: Unsupervised Learning of Optical Flow with a Bidirectional Census Loss
Stars: ✭ 239 (+602.94%)
Mutual labels:  optical-flow
vision-based estimations
Vision-based Robot 3D Pose and Velocities Estimations
Stars: ✭ 32 (-5.88%)
Mutual labels:  optical-flow
EPC
Every Pixel Counts ++: Joint Learning of Geometry and Motion with 3D Holistic Understanding
Stars: ✭ 27 (-20.59%)
Mutual labels:  optical-flow
BridgeDepthFlow
Bridging Stereo Matching and Optical Flow via Spatiotemporal Correspondence, CVPR 2019
Stars: ✭ 114 (+235.29%)
Mutual labels:  optical-flow
CrowdFlow
Optical Flow Dataset and Benchmark for Visual Crowd Analysis
Stars: ✭ 87 (+155.88%)
Mutual labels:  optical-flow
GuidedNet
Caffe implementation for "Guided Optical Flow Learning"
Stars: ✭ 28 (-17.65%)
Mutual labels:  optical-flow
flowattack
Attacking Optical Flow (ICCV 2019)
Stars: ✭ 58 (+70.59%)
Mutual labels:  optical-flow
Optical-Flow-based-Obstacle-Avoidance
Image based obstacle avoidance using optical flow
Stars: ✭ 24 (-29.41%)
Mutual labels:  optical-flow
PCLNet
Unsupervised Learning for Optical Flow Estimation Using Pyramid Convolution LSTM.
Stars: ✭ 29 (-14.71%)
Mutual labels:  optical-flow
Maskflownet
[CVPR 2020, Oral] MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Mask
Stars: ✭ 242 (+611.76%)
Mutual labels:  optical-flow
CorrelationLayer
Pure Pytorch implementation of Correlation Layer that commonly used in learning based optical flow estimator
Stars: ✭ 22 (-35.29%)
Mutual labels:  optical-flow
Optical-Flow-GPU-Docker
Compute dense optical flow using TV-L1 algorithm with NVIDIA GPU acceleration.
Stars: ✭ 48 (+41.18%)
Mutual labels:  optical-flow
Face-Detection-and-Tracking
Face Detection and tracking using CamShift, Kalman Filter, Optical Flow
Stars: ✭ 30 (-11.76%)
Mutual labels:  optical-flow
deepOF
TensorFlow implementation for "Guided Optical Flow Learning"
Stars: ✭ 26 (-23.53%)
Mutual labels:  optical-flow

EPPM

Code by Linchao Bao ([email protected])

About

EPPM is an implementation of the optical flow algorithm presented in the paper "Fast Edge-Preserving PatchMatch for Large Displacement Optical Flow" in CVPR 2014.

The source code is different from the original paper in the following ways:

  1. The self-similarity propagation scheme proposed in the paper is not used in this implementation. Instead, the implementation uses pixel skipping scheme to reduce patch matching computation.

  2. The parameter setting are not the same as that for benchmarking in the paper.

Compiling

The code relies on NVIDIA CUDA (version > 5.0) and only works on NVIDIA GPUs. To compile the code, run the following command:

mkdir build
cd build
cmake ..
make

Citing

Please cite the following paper if you find the code useful:

@inproceedings{bao2014cvpreppm,
  title={Fast Edge-Preserving PatchMatch for Large Displacement Optical Flow},
  author={Bao, Linchao and Yang, Qingxiong and Jin, Hailin},
  booktitle={IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2014},
  pages={3534-3541}, 
  organization={IEEE}
}

License & Disclaimer

The code is released under an MIT license and is developed for academic purpose. Please use the code in your own risk.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].