Skip to content

Latest commit

 

History

History
21 lines (14 loc) · 1010 Bytes

precompiled_modelartifacts.md

File metadata and controls

21 lines (14 loc) · 1010 Bytes

Pre-complied / pre-imported model artifacts

This package provides Pre-Compiled Model Artifacts for several Deep Neural Network models.

These artifacts can be used for inference in multiple ways:

For inference/benchmark in edgeai-benchmark

In EdgeAI SDKs of the supported SoCs PROCESSOR-SDK-LINUX-SK-TDA4VM

Note: The tasks performed by the pre-compiled model artifacts can be identified using the following fields in their names:

  • cl: Image Classification Models
  • od: Object Detection Models (COCO Object detection, Face detection)
  • ss: Semantic Segmentation Models
  • kd: Keypoint Detection Models (Human Pose Estimation)
  • de: Depth Estimation
  • lidar-3dod: 3D Object detection using LIDAR input
  • 6d-pose: 6D Pose estimation of objects (3-Translation, 3-Rotation)