Scribd Logo
Upload

Welcome to Scribd!

  • 39 views

    Super Resolution in Opencv

    Uploaded by

    YingHao Tiew
    The document discusses super resolution in OpenCV using deep learning methods. It provides steps to use a pre-trained super resolution model in OpenCV, including: 1) Installing OpenCV with contrib modules containing the dnn_superres interface, 2) Downloading a pre-trained model separately, as they are not included in OpenCV, 3) Upscaling an image using the model. Pre-trained models available can increase image resolution by 2x or more and are based on popular super resolution papers.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    Super Resolution in Opencv

    Uploaded by

    YingHao Tiew
    39 views2 pages
    The document discusses super resolution in OpenCV using deep learning methods. It provides steps to use a pre-trained super resolution model in OpenCV, including: 1) Installing OpenCV with contrib modules containing the dnn_superres interface, 2) Downloading a pre-trained model separately, as they are not included in OpenCV, 3) Upscaling an image using the model. Pre-trained models available can increase image resolution by 2x or more and are based on popular super resolution papers.

    Original Title

    Upscaling

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses super resolution in OpenCV using deep learning methods. It provides steps to use a pre-trained super resolution model in OpenCV, including: 1) Installing OpenCV with contrib modules containing the dnn_superres interface, 2) Downloading a pre-trained model separately, as they are not included in OpenCV, 3) Upscaling an image using the model. Pre-trained models available can increase image resolution by 2x or more and are based on popular super resolution papers.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    39 views2 pages

    Super Resolution in Opencv

    Uploaded by

    YingHao Tiew
    The document discusses super resolution in OpenCV using deep learning methods. It provides steps to use a pre-trained super resolution model in OpenCV, including: 1) Installing OpenCV with contrib modules containing the dnn_superres interface, 2) Downloading a pre-trained model separately, as they are not included in OpenCV, 3) Upscaling an image using the model. Pre-trained models available can increase image resolution by 2x or more and are based on popular super resolution papers.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    of 2

    Super Resolution with Deep Learning

    https://www.pyimagesearch.com/2020/11/09/opencv-super-resolution-with-deep-learning/

    https://towardsdatascience.com/deep-learning-based-super-resolution-with-opencv-4fd736678066

    https://learnopencv.com/super-resolution-in-opencv/ (Model can be downloaded from this link)

    https://docs.opencv.org/4.5.2/dc/d69/tutorial_dnn_superres_benchmark.html

    https://docs.opencv.org/4.5.2/d8/df8/tutorial_table_of_content_dnn_superres.html

    Super Resolution in OpenCV

    OpenCV is an open-source computer vision library that has an extensive collection of

    great algorithms. Since one of the latest mergers, OpenCV contains an easy-to-use

    interface for implementing Super Resolution (SR) based on deep learning methods.

    The interface contains pre-trained models that can be used for inference very easily

    and efficiently. In this post, I will explain what it can do and show step-by-step how to

    use it. So far, it works with C++ and Python.

    NOTE: You don’t need any knowledge about SR to use this code. I am running this on

    an Ubuntu 16.04 machine, but it works for any other distributions (it also works on

    Windows).

    Steps

     1. Installing OpenCV with the contrib modules (containing dnn_superres).

     2. Downloading the pre-trained models.

     3. Upscaling the image.

    1. Installing OpenCV with the contrib modules.

    Our first step is to install OpenCV. Some features are released later, so be

    careful with the version: 4.2.0 for C++, 4.3.0 adds Python wrap, 4.4.0
    adds GPU inference. You can follow the instructions from the opencv

    documentation. It is very important that you also install the contrib modules,

    because that is where the SR interface code resides. You can choose to install all the

    contrib modules or just the necessary SR one. The interface or module we will use is

    called dnn_superres (dnn stands for Deep Neural Network; superres for Super

    Resolution).

    2. Downloading a pre-trained model.

    We need to download the pre-trained models separately, because the OpenCV code-

    base does not contain them. The reason is some of the models are quite big. So if they

    were in the standard code-base, people who might not need them still would have to

    download them. There are several models to choose from (see section Models in this

    article). All of them are implementations of popular SR papers. For now, let’s choose a

    small model. You can download it here. This pre-trained model can increase the

    resolution of an image by 2. For more models and their explanations, see section

    Models in this post.

    You might also like