Scale the input by applying one of the super-resolution methods based on convolutional neural networks. Supported models:

• Super-Resolution Convolutional Neural Network model (SRCNN). See https://arxiv.org/abs/1501.00092.

• Efficient Sub-Pixel Convolutional Neural Network model (ESPCN). See https://arxiv.org/abs/1609.05158.

Training scripts as well as scripts for model file (.pb) saving can be found at https://github.com/XueweiMeng/sr/tree/sr_dnn_native. Original repository is at https://github.com/HighVoltageRocknRoll/sr.git.

The filter accepts the following options:

dnn_backend

Specify which DNN backend to use for model loading and execution. This option accepts the following values:

tensorflow

TensorFlow backend. To enable this backend you need to install the TensorFlow for C library (see https://www.tensorflow.org/install/lang_c) and configure FFmpeg with --enable-libtensorflow

model

Set path to model file specifying network architecture and its parameters. Note that different backends use different file formats. TensorFlow, OpenVINO backend can load files for only its format.

scale_factor

Set scale factor for SRCNN model. Allowed values are 2, 3 and 4. Default value is 2. Scale factor is necessary for SRCNN model, because it accepts input upscaled using bicubic upscaling with proper scale factor.

To get full functionality (such as async execution), please use the dnn_processing filter.