vitis::ai::TFSSD - 2.0 English

Vitis AI Library User Guide (UG1354)

Document ID
UG1354
Release Date
2022-01-20
Version
2.0 English
Base class for detecting 90 objects of the COCO dataset.

Input is an image (cv:Mat).

Output is a struct of detection results, named TFSSDResult.

Sample code :

  Mat img = cv::imread("sample_tfssd.jpg");
  auto tfssd = vitis::ai::TFSSD::create("ssd_resnet_50_fpn_coco_tf",true);
  auto results = tfssd->run(img);
  for(const auto &r : results.bboxes){
     auto label = r.label;
     auto x = r.x * img.cols;
     auto y = r.y * img.rows;
     auto width = r.width * img.cols;
     auto height = r.height * img.rows;
     auto score = r.score;
     std::cout << "RESULT: " << label << "\t" << x << "\t" << y << "\t" <<
width
        << "\t" << height << "\t" << score << std::endl;
  }

Display of the model results:

Figure 1. detection result
Image sample_tfssd_result.jpg

Quick Function Reference

The following table lists all the functions defined in the vitis::ai::TFSSD class:

Table 1. Quick Function Reference
Type Name Arguments
std::unique_ptr< TFSSD > create
  • const std::string & model_name
  • bool need_preprocess
vitis::ai::TFSSDResult run
  • const cv::Mat & img
std::vector< vitis::ai::TFSSDResult > run
  • const std::vector< cv::Mat > & imgs
int getInputWidth
  • void
int getInputHeight
  • void
size_t get_input_batch
  • void