Vision-based AI Use Cases

Code samples and explanations for detecting facial expressions

Detecting and localizing human faces paves the way for facial expression recognition (FER), which has many applications ranging from entertainment (face filters) to driver safety (emotion analysis).

With the Qualcomm® Neural Processing SDK for AI, developers can implement trained neural networks on Snapdragon® mobile platforms. The following pages describe vision-based AI use cases for FER, including categorizing facial expressions and testing apps.

Real-time Facial Detection and Validation

Using computer vision and AI to recognize facial expressions and enhance vehicle safety

Functional Testing of Vision-based ML Applications

Test scenarios and test cases for an ML model and app to detect drowsiness

Facial Keypoint Detection

Using a CNN to detect facial keypoints and use them in applications with face filters

Facial Expression Recognition — Part 1: Solution Pipeline on Ubuntu

Pre-processing the data set, training the model and running inference on the desktop

Facial Expression Recognition — Part 2: Solution Pipeline on Android

Pre-processing the data set, training the model and running inference on the desktop

Performance Analysis Using Benchmarking Tools

Tools included in Qualcomm Neural Processing SDK