vitis::ai::MultiTaskv3 - 1.4.1 English

Vitis AI Library User Guide (UG1354)

Document ID
UG1354
Release Date
2021-12-11
Version
1.4.1 English
Multitaskv3 Network Type , declaration multitask Network.

number of classes label: 0 name: "car" label: 1 name: "sign" label: 2 name: "person"

Base class for ADAS MuiltTask from an image (cv::Mat).

Input an image (cv::Mat).

Output is a struct of MultiTaskv3Result include segmentation results, detection results and vehicle towards;

Sample code:

auto det = vitis::ai::MultiTaskv3::create("multi_task");
auto image = cv::imread("sample_multitaskv3.jpg");
auto result = det->run_8UC3(image);
cv::imwrite("sample_multitaskv3_result.jpg",result.segmentation);
cv::imwrite("sample_multitaskv3_result.jpg",result.depth);

Display of the model results:

Figure 1. result image
Image sample_multitaskv3_result.jpg

Quick Function Reference

The following table lists all the functions defined in the vitis::ai::MultiTaskv3 class:

Table 1. Quick Function Reference
Type Name Arguments
std::unique_ptr< MultiTaskv3 > create
  • const std::string & model_name
  • bool need_preprocess
int getInputWidth
  • void
int getInputHeight
  • void
size_t get_input_batch
  • void
MultiTaskv3Result run_8UC1
  • const cv::Mat & image
std::vector< MultiTaskv3Result > run_8UC1
  • const std::vector< cv::Mat > & images
MultiTaskv3Result run_8UC3
  • const cv::Mat & image
std::vector< MultiTaskv3Result > run_8UC3
  • const std::vector< cv::Mat > & images