VPI-3.1.5

VPI-3.1.5 (post-release updates)

Documentation

• Fixed OFA support on Orin in maximize clocks script.
• Removed support for Xavier devices from the maximize clocks script.

VPI-3.1.5

VPI-3.1.5 is the first production release of VPI-3.1 branch. It contains several new algorithms and bug fixes.

Documentation

Link to documentation for release VPI-3.0

Release Highlights

• Median Filter :
  • Up to 4x performance optimization on VPI_BACKEND_CUDA.
• Erode :
  • Added support for VPI_BACKEND_PVA. This includes support for:
    • Filter sizes 3x3 and 5x5.
    • Image formats VPI_IMAGE_FORMAT_U8, VPI_IMAGE_FORMAT_S8, VPI_IMAGE_FORMAT_U16, VPI_IMAGE_FORMAT_S16 .
    • Border types VPI_BORDER_ZERO and VPI_BORDER_CLAMP .
    • Filter shapes ALL_TRUE, ALL_FALSE and CROSS (Only for 3x3).
• Dilate :
  • Added support for VPI_BACKEND_PVA. This includes support for:
    • Filter sizes 3x3 and 5x5.
    • Image formats VPI_IMAGE_FORMAT_U8, VPI_IMAGE_FORMAT_S8, VPI_IMAGE_FORMAT_U16, VPI_IMAGE_FORMAT_S16 .
    • Border types VPI_BORDER_ZERO and VPI_BORDER_CLAMP .
    • Filter shapes ALL_TRUE, ALL_FALSE and CROSS (Only for 3x3).

Other Updates

• Crop Scaler :
  • Added support for VPI_IMAGE_FORMAT_BGR8p.
  • Max frames increased to 128.
• ORB feature detector :
  • Functional changes to the algorithm.
    • Now works with pyramidal keypoints, meaning it extracts keypoints with octave information and uses them to compute descriptor at appropriate octave image (pyramid level).
• Convert Image Format :
  • Added support for conversion from VPI_IMAGE_FORMAT_2S16_BL to VPI_IMAGE_FORMAT_2S16_PL and vice versa for VPI_BACKEND_VIC
• Gaussian Pyramid Generator :
  • Added support for VPI_BORDER_REFLECT and VPI_BORDER_MIRROR for VPI_BACKEND_CUDA.
• Laplacian Pyramid Generator :
  • Added support for VPI_BORDER_REFLECT and VPI_BORDER_MIRROR for VPI_BACKEND_CUDA.
• DCF Tracker :
  • Increased max sequences to 33 for VPI_BACKEND_PVA.
  • Increased max sequences to 1024 for VPI_BACKEND_CUDA.

Selected Bug Fixes

• Fixed segfault with python bindings when importing modules like torch after vpi.

Known Issues

• Host images wrapped into VPIImages using vpiImageCreateWrapper might impact performance when using them with algorithms running on the CUDA backend. You should avoid wrappers in this case, and use VPIImages allocated with vpiImageCreate instead.
• Performance might be affected when using CUDA images wrapped into VPIImages using vpiImageCreateWrapper in algorithms running in PVA, VIC and/or OFA. User should avoid using wrappers in this case, preferring to use VPIImages allocated with vpiImageCreate.
• Harris Corner Detector result scores/positions might differ among backends.
• Stereo Disparity Estimator
  • The confidence map generated by OFA+PVA+VIC backend might have some negligible differences with respect to other backends.
• Performance might be affected when using Dense Optical Flow on python due re-creating the payload at every call.

• Attention

  Algorithms running on PVA backend won't work inside a docker container. Submission calls will return error VPI_ERROR_INTERNAL .