超分辨率相关资源大列表-2

本文收藏自Github,仅供学习交流所用。

Video Super Resolution

A collection of state-of-the-art video or single-image super-resolution architectures, reimplemented in tensorflow.

Pretrained weights is uploading now.

Several referenced PyTorch implementations are also included now.

Quick Link: Installation Getting Started

Network list and reference (Updating)

The hyperlink directs to paper site, follows the official codes if the authors open sources.

All these models are implemented in ONE framework.

ModelPublishedCode*VSR (TF)**VSR (Torch)KeywordsPretrained
SRCNNECCV14-, KerasY-Kaiming
RAISRarXiv---Google, Pixel 3 
ESPCNCVPR16-, KerasYYReal time
VDSRCVPR16-Y-Deep, Residual
DRCNCVPR16-Y-Recurrent 
DRRNCVPR17CaffePyTorchY-Recurrent 
LapSRNCVPR17MatlabY-Huber loss 
EDSRCVPR17-YYNTIRE17 Champion
SRGANCVPR17-Y-1st proposed GAN 
VESPCNCVPR17-YYVideoSR
MemNetICCV17CaffeY-  
SRDenseNetICCV17-, PyTorchY-Dense
SPMCICCV17Tensorflow--VideoSR 
DnCNNTIP17MatlabY-Denoise
DCSCNarXivTensorflowY-  
IDNCVPR18CaffeY-Fast
RDNCVPR18TorchY-Deep, BI-BD-DN 
SRMDCVPR18MatlabT-Denoise/Deblur/SR 
DBPNCVPR18PyTorchYYNTIRE18 Champion
ZSSRCVPR18Tensorflow--Zero-shot 
FRVSRCVPR18PDFTYVideoSR 
DUFCVPR18TensorflowT-VideoSR 
CARNECCV18PyTorchYYFast
RCANECCV18PyTorchYYDeep, BI-BD-DN 
MSRNECCV18PyTorchYY 
SRFeatECCV18TensorflowYTGAN 
NLRNNIPS18TensorflowT-Non-local, Recurrent 
SRCliqueNetNIPS18---Wavelet 
FFDNetTIP18MatlabY-Conditional denoise 
CBDNetarXivMatlabT-Blind-denoise 
ESRGANECCVW18PyTorch-Y1st place PIRM 2018 
TecoGANarXivTensorflow-TVideoSR GAN 
RBPNCVPR19PyTorch--VideoSR 

*The 1st repo is by paper author.

**Y: included; -: not included; T: under-testing.

You can download pre-trained weights through prepare_data, or visit the hyperlink at .

Link of datasets

(please contact me if any of links offend you or any one disabled)

NameUsage#SiteComments
SET5Test5downloadjbhuang0604
SET14Test14downloadjbhuang0604
SunHay80Test80downloadjbhuang0604
Urban100Test100downloadjbhuang0604
VID4Test4download4 videos
BSD100Train300downloadjbhuang0604
BSD300Train/Val300download-
BSD500Train/Val500download-
91-ImageTrain91downloadYang
DIV2KTrain/Val900websiteNTIRE17
WaterlooTrain4741website-
MCL-VTrain12website12 videos
GOPROTrain/Val33website33 videos, deblur
CelebATrain202599websiteHuman faces
SintelTrain/Val35websiteOptical flow
FlyingChairsTrain22872websiteOptical flow
DNDTest50websiteReal noisy photos
RENOIRTrain120websiteReal noisy photos
NCTest60websiteNoisy photos
SIDD(M)Train/Val200websiteNTIRE 2019 Real Denoise
RSRTrain/Val80downloadNTIRE 2019 Real SR
Vimeo-90kTrain/Test89800website90k HQ videos

Other open datasets: Kaggle ImageNet COCO

VSR package

This package offers a training and data processing framework based on TF. What I made is a simple, easy-to-use framework without lots of encapulations and abstractions. Moreover, VSR can handle raw NV12/YUV as well as a sequence of images as inputs.

Install

  1. Prepare proper tensorflow and pytorch(optional). For example, GPU and CUDA10.0 (recommend to use conda):

    conda install tensorflow-gpu==1.12.0
    # optional
    # conda install pytorch==1.0.1
  2. Install VSR package

    # For someone see this doc online
    # git clone https://github.com/loseall/VideoSuperResolution && cd VideoSuperResolution
    pip install -e .

Getting Started

  1. Download pre-trained weights and (optinal) training datasets. For instance, let's begin with VESPCN and vid4 test data:

    python prepare_data.py --filter vespcn vid4
  2. Evaluate

    cd Train
    python run.py --model vespcn --test vid4
  3. Train

    python prepare_data.py --filter mcl-v
    cd Train
    python run.py --model vespcn --dataset mcl-v --memory_limit 1GB --epochs 100

OK, that's all you need. For more details, use --help to get more information.

To train/test/infer any other model in VSR.Models, please see README. To write and train your own model via VSR, please see Docs.

  • 0
    点赞
  • 3
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值