Overview

The TAO Toolkit Computer Vision (CV) Inference Pipeline is a C++ based SDK that provides APIs to build applications from inferences from purpose-built pre-trained AI models. The underlying framework provides a foundation to build multimodal applications. For example, the Gaze Estimation sample application requires the combination of Face Detection and Facial Landmarks (Fiducial Keypoints) Estimation.

The TAO Computer Vision Inference Pipeline is made up of three key components:

The purpose-built AI models that are supported by this Inference Pipeline are as follows:

Model Deployment

Model

Model output format

Prunable

INT8

Compatible with TAO Toolkit CV Inference Pipeline

BodyPoseNet

Encrypted ONNX

Yes

Yes

Yes

EmotionNet

Encrypted ONNX

No

No

Yes

FaceNet

Encrypted UFF

Yes

Yes

Yes

FPENet

Encrypted ONNX

No

Yes

Yes

GazeNet

Encrypted ONNX

No

No

Yes

GestureNet

Encrypted ONNX

No

Yes

Yes

HeartRateNet

Encrypted ONNX

No

No

Yes

Users can retrain supported TAO networks, drop the optimized TensorRT models into the NVIDIA Triton Inference Server, and build their own AI applications and use-cases using the TAO Toolkit CV API.

The deployment of the Inference Pipeline is managed by the TAO Toolkit CV Quick Start Scripts, which are bash scripts that pull/start relevant containers, compile TAO models, and start the Triton Server.