ClementPinard / Pytorch Correlation Extension
Licence: mit
Custom implementation of Corrleation Module
Stars: ✭ 268
Programming Languages
python
139335 projects - #7 most used programming language
Labels
Projects that are alternatives of or similar to Pytorch Correlation Extension
LSUV-keras
Simple implementation of the LSUV initialization in keras
Stars: ✭ 65 (-75.75%)
Mutual labels: deeplearning
Deep-Learning
It contains the coursework and the practice I have done while learning Deep Learning.🚀 👨💻💥 🚩🌈
Stars: ✭ 21 (-92.16%)
Mutual labels: deeplearning
Yolov5-deepsort-driverDistracted-driving-behavior-detection
基于深度学习的驾驶员分心驾驶行为(疲劳+危险行为)预警系统使用YOLOv5+Deepsort实现驾驶员的危险驾驶行为的预警监测
Stars: ✭ 107 (-60.07%)
Mutual labels: deeplearning
Image-Forgery-using-Deep-Learning
Image Forgery Detection using Deep Learning, implemented in PyTorch.
Stars: ✭ 46 (-82.84%)
Mutual labels: deeplearning
battery-rul-estimation
Remaining Useful Life (RUL) estimation of Lithium-ion batteries using deep LSTMs
Stars: ✭ 25 (-90.67%)
Mutual labels: deeplearning
recurrent-defocus-deblurring-synth-dual-pixel
Reference github repository for the paper "Learning to Reduce Defocus Blur by Realistically Modeling Dual-Pixel Data". We propose a procedure to generate realistic DP data synthetically. Our synthesis approach mimics the optical image formation found on DP sensors and can be applied to virtual scenes rendered with standard computer software. Lev…
Stars: ✭ 30 (-88.81%)
Mutual labels: deeplearning
Jejunet
Real-Time Video Segmentation on Mobile Devices with DeepLab V3+, MobileNet V2. Worked on the project in 🏝 Jeju island
Stars: ✭ 258 (-3.73%)
Mutual labels: deeplearning
Multi-Face-Comparison
This repo is meant for backend API for face comparision and computer vision. It is built on python flask framework
Stars: ✭ 20 (-92.54%)
Mutual labels: deeplearning
dilation-keras
Multi-Scale Context Aggregation by Dilated Convolutions in Keras.
Stars: ✭ 72 (-73.13%)
Mutual labels: deeplearning
Ensemble-Pytorch
A unified ensemble framework for PyTorch to improve the performance and robustness of your deep learning model.
Stars: ✭ 407 (+51.87%)
Mutual labels: deeplearning
WSCNNTDSaliency
[BMVC17] Weakly Supervised Saliency Detection with A Category-Driven Map Generator
Stars: ✭ 19 (-92.91%)
Mutual labels: deeplearning
ohsome2label
Historical OpenStreetMap Objects to Machine Learning Training Samples
Stars: ✭ 33 (-87.69%)
Mutual labels: deeplearning
deepstreet
Traffic Sign Recognition - Fine tuning VGG16 + GTSRB
Stars: ✭ 60 (-77.61%)
Mutual labels: deeplearning
Similarity-Adaptive-Deep-Hashing
Unsupervised Deep Hashing with Similarity-Adaptive and Discrete Optimization (TPAMI2018)
Stars: ✭ 18 (-93.28%)
Mutual labels: deeplearning
Fixmatch Pytorch
Unofficial PyTorch implementation of "FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence"
Stars: ✭ 259 (-3.36%)
Mutual labels: deeplearning
Data-Analysis
Different types of data analytics projects : EDA, PDA, DDA, TSA and much more.....
Stars: ✭ 22 (-91.79%)
Mutual labels: deeplearning
HistoGAN
Reference code for the paper HistoGAN: Controlling Colors of GAN-Generated and Real Images via Color Histograms (CVPR 2021).
Stars: ✭ 158 (-41.04%)
Mutual labels: deeplearning
Pytorch Correlation module
this is a custom C++/Cuda implementation of Correlation module, used e.g. in FlowNetC
This tutorial was used as a basis for implementation, as well as NVIDIA's cuda code
- Build and Install C++ and CUDA extensions by executing
python setup.py install
, - Benchmark C++ vs. CUDA by running
python benchmark.py {cpu, cuda}
, - Run gradient checks on the code by running
python grad_check.py --backend {cpu, cuda}
.
Requirements
This module is expected to compile for Pytorch 1.6
.
Installation
this module is available on pip
pip install spatial-correlation-sampler
For a cpu-only version, you can install from source with
python setup_cpu.py install
Known Problems
This module needs compatible gcc version and CUDA to be compiled. Namely, CUDA 9.1 and below will need gcc5, while CUDA 9.2 and 10.0 will need gcc7 See this issue for more information
Usage
API has a few difference with NVIDIA's module
- output is now a 5D tensor, which reflects the shifts horizontal and vertical.
input (B x C x H x W) -> output (B x PatchH x PatchW x oH x oW)
- Output sizes
oH
andoW
are no longer dependant of patch size, but only of kernel size and padding - Patch size
patch_size
is now the whole patch, and not only the radii. -
stride1
is nowstride
andstride2
isdilation_patch
, which behave like dilated convolutions - equivalent
max_displacement
is thendilation_patch * (patch_size - 1) / 2
. -
dilation
is a new parameter, it acts the same way as dilated convolution regarding the correlation kernel - to get the right parameters for FlowNetC, you would have
kernel_size=1
patch_size=21,
stride=1,
padding=0,
dilation=1
dilation_patch=2
Example
import torch
from spatial_correlation_sampler import SpatialCorrelationSampler,
device = "cuda"
batch_size = 1
channel = 1
H = 10
W = 10
dtype = torch.float32
input1 = torch.randint(1, 4, (batch_size, channel, H, W), dtype=dtype, device=device, requires_grad=True)
input2 = torch.randint_like(input1, 1, 4).requires_grad_(True)
#You can either use the function or the module. Note that the module doesn't contain any parameter tensor.
#function
out = spatial_correlation_sample(input1,
input2,
kernel_size=3,
patch_size=1,
stride=2,
padding=0,
dilation=2,
dilation_patch=1)
#module
correlation_sampler = SpatialCorrelationSampler(
kernel_size=3,
patch_size=1,
stride=2,
padding=0,
dilation=2,
dilation_patch=1)
out = correlation_sampler(input1, input2)
Benchmark
- default parameters are from
benchmark.py
, FlowNetC parameters are same as use inFlowNetC
with a batch size of 4, described in this paper, implemented here and here. - Feel free to file an issue to add entries to this with your hardware !
CUDA Benchmark
- See here for a benchmark script working with NVIDIA's code, and Pytorch.
- Benchmark are launched with environment variable
CUDA_LAUNCH_BLOCKING
set to1
. - Only
float32
is benchmarked. - FlowNetC correlation parameters where launched with the following command:
CUDA_LAUNCH_BLOCKING=1 python benchmark.py --scale ms -k1 --patch 21 -s1 -p0 --patch_dilation 2 -b4 --height 48 --width 64 -c256 cuda -d float
CUDA_LAUNCH_BLOCKING=1 python NV_correlation_benchmark.py --scale ms -k1 --patch 21 -s1 -p0 --patch_dilation 2 -b4 --height 48 --width 64 -c256
implementation | Correlation parameters | device | pass | min time | avg time |
---|---|---|---|---|---|
ours | default | 980 GTX | forward | 5.745 ms | 5.851 ms |
ours | default | 980 GTX | backward | 77.694 ms | 77.957 ms |
NVIDIA | default | 980 GTX | forward | 13.779 ms | 13.853 ms |
NVIDIA | default | 980 GTX | backward | 73.383 ms | 73.708 ms |
ours | FlowNetC | 980 GTX | forward | 26.102 ms | 26.179 ms |
ours | FlowNetC | 980 GTX | backward | 208.091 ms | 208.510 ms |
NVIDIA | FlowNetC | 980 GTX | forward | 35.363 ms | 35.550 ms |
NVIDIA | FlowNetC | 980 GTX | backward | 283.748 ms | 284.346 ms |
Notes
- The overhead of our implementation regarding
kernel_size
> 1 during backward needs some investigation, feel free to dive in the code to improve it ! - The backward pass of NVIDIA is not entirely correct when stride1 > 1 and kernel_size > 1, because not everything is computed, see here.
CPU Benchmark
- No other implementation is avalaible on CPU.
- It is obviously not recommended to run it on CPU if you have a GPU.
Correlation parameters | device | pass | min time | avg time |
---|---|---|---|---|
default | E5-2630 v3 @ 2.40GHz | forward | 159.616 ms | 188.727 ms |
default | E5-2630 v3 @ 2.40GHz | backward | 282.641 ms | 294.194 ms |
FlowNetC | E5-2630 v3 @ 2.40GHz | forward | 2.138 s | 2.144 s |
FlowNetC | E5-2630 v3 @ 2.40GHz | backward | 7.006 s | 7.075 s |
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].