classify

• “classify” filter doesn’t work independently, it can only be inserted to the pipeline after “detect” filter. Multiple “classify” filters can be inserted in a cascade, to load and execute different classification models.
• “classify” filter executes inference on the objects detected by “detect” filter. There may be zero or multiple objects detected. For every objects, the filter crops, scales and executes CSC (to BGRP format generally which is supported by the models) to the objects and finally executes the inference. Similar to “detect” filter, the inference results are kept in AVFrameSideData instances in AVFrame.

Filter classify
  Image Inference classify filter.
    Inputs:
       #0: default (video)
    Outputs:
       #0: default (video)
inference_classify AVOptions:
  dnn_backend       <flags>      ..FV..... DNN backend for model execution (default 1)
  model             <string>     ..FV..... path to model file for network
  model_proc        <string>     ..FV..... model preproc and postproc
  object_class      <string>     ..FV..... objective class
  device            <string>     ..FV..... running on device name
  configs           <string>     ..FV..... configurations to backend
  interval          <int>        ..FV..... detect every Nth frame (from 1 to 1024) (default 1)
  nireq             <int>        ..FV..... inference request number (from 1 to 128) (default 1)
  batch_size        <int>        ..FV..... batch size per infer (from 1 to 1000) (default 1)
  threshold         <float>      ..FV..... threshod to filter output data (from 0 to 1) (default 0.5)
  async_preproc     <boolean>    ..FV..... do asynchronous preproc in inference backend (default false)