FAQ

1. Is VVAS open source?

Answer: Yes.

2. What platforms and OS are compatible with VVAS?

Answer: VVAS is tested on PetaLinux for embedded platforms and Ubuntu 20.04 for PCIe based platforms. For more information about supported platforms, refer to Software & Hardware Requirements.

3. Which Vitis AI classes are supported with VVAS SDK?

Answer: Below classes are supported.

  • YOLOV3

  • FACEDETECT

  • CLASSIFICATION

  • SSD

  • REFINEDET

  • TFSSD

  • YOLOV2

  • VEHICLECLASSIFICATION

  • REID

  • SEGMENTATION

  • PLATEDETECT

  • PLATENUM

  • POSEDETECT

  • BCC

  • EFFICIENTDETD2

  • FACEFEATURE

  • FACELANDMARK

  • ROADLINE

  • ULTRAFAST

  • RAWTENSOR

For more information refer to Supporting Deep Learning Models

4. How do I enable models that are not officially supported?

Answer: Refer to How to the page Supporting Deep Learning Models

5. Does VVAS compatible with lower versions of Vitis AI tools, such as Vitis-AI v2.5?

Answer: No, it has dependencies on Vitis AI 3.0.

6. How can I change the model in the pipeline?

Answer: The model name to be used for inference has to be provided in the JSON file for vvas_xinfer plug-in. For more details, see vvas_xinfer.

7. Can the model be changed dynamically?

Answer: while a pipeline is running, the model parameters cannot be modified. To change the model’s parameters, stop the running pipeline, and then update the JSON file and then re-start the pipeline.

8. What types of input streams does current VVAS SDK support?

Answer:

  • H.264, H.265 encoded video streams

  • Raw video frames in NV12, BGR/RGB formats

9. Whether VVAS SDK supports input of RTSP streaming?

Answer: Receiving RTSP stream is supported by an open source plugin.

10. How do I develop software acceleration libraries?

Answer: Refer to Acceleration Acceleration Software Library Development Guide

11. How to debug VVAS application if there are any issues?

Answer: Refer to the Debugging

12. How do I compile and prune the model to be used?

Answer: Refer to Vitis AI 3.0 documentation.

12. How can we communicate issues while using VVAS SDK ?

Answer: Please visit technical support page https://support.xilinx.com/s/article/How-do-I-get-support-for-Xilinx-technical-issues?language=en_US

13. What if I cannot find the information that i am looking for?

Answer: Contact vvas_discuss@amd.com.