GoPoint
GoPoint is a user-friendly application that allows the user to launch preselected demonstrations included in the NXP provided BSP and follows the quarterly release roadmap for BSP
- How to launch GoPoint
GoPoint Demo On FRDM-IMX93 Board
Since FRDM-IMX93 board’s BSP is based on standard BSP release, GoPoint is included in FRDM-IMX93 Yocto build by default.
List of 9 demos available on FRDM-IMX93 Board:
- Image Classification
- Object Detection
- Selfie Segmenter
- i.MX Smart Fitness
- DMS (Driver Monitor System)
- ML Benchmark
- Video Test
- i.MX Smart Kitchen
- i.MX E-Bike VIT
Image Classification Demo
Image classification is a ML task that attempts to comprehend an entire image as a whole. The goal is to classify the image by assigning it to a specific label. Typically, it refers to images in which only one object appears and is analyzed. This example is using NNStreamer.
Object Detection Demo
Object detection is the ML task that detects instances of objects of a certain class within an image. A bounding box and a class label are found for each detected object. This example is using NNStreamer.
Selfie Segmenter Demo
Selfie Segmenter showcases the ML capabilities of i.MX 93 by using the NPU to accelerate an instance segmentation model. This model lets you segment the portrait of a person and can be used to replace or modify the background of an image. This example is using NNStreamer.
i.MX Smart Fitness Demo
i.MX Smart Fitness showcases the i.MX' Machine Learning capabilities by using an NPU to accelerate two Deep Learning vision-based models. Together, these models detect a person present in the scene and predict 33 3D-keypoints to generate a complete body landmark, known as pose estimation. From the pose estimation, the application tracks the 'squats' fitness exercise.
DMS (Driver Monitor System) Demo
This application showcases the capability of implementing DMS on i.MX 93 platform, and the performance boost brought by Neural Processing Unit (NPU). DMS uses four ML models in total to achieve face detection, capturing face landmark and iris landmark, smoking detection and calling detection.
ML Benchmark Demo
This example is based on benchmark_model tool in Tensorflow Lite framework, which allows to easily compare the performance of TensorFlow Lite models running on CPU (Cortex-A) and NPU.
Video Test Demo
This is a simple demo that allows users to play back video captured on a camera or a test source. It’s based on gstreamer pipeline.
i.MX Smart Kitchen Demo
i.MX Smart Kitchen showcases the Multimedia capabilities of i.MX to emulate an interactive kitchen through a GUI controlled by voice commands. The GUI is based on LVGL (Little Versatile Graphic Library) and NXP's Voice Intelligent Technology (VIT) supports the voice commands.
- Usage: Keyword + command
i.MX E-Bike VIT Demo
i.MX E-Bike VIT showcases the Multimedia capabilities of i.MX to emulate an interactive ebike through a GUI controlled by voice commands. The GUI is based on LVGL (Little Versatile Graphic Library) and NXP's Voice Intelligent Technology (VIT) supports the voice commands.
- Usage: Keyword + command
Useful Link
- GoPoint User Guide: https://www.nxp.com/webapp/Download?colCode=GPNTUG
- GoPoint repo: https://github.com/nxp-imx-support/nxp-demo-experience-demos-list/tree/lf-6.6.36_2.1.0 (Including source code of demo: Selfie Segmenter, DMS, ML benchmark, Video test)
- Image Classification/Object Detection: https://github.com/nxp-imx/eiq-example/tree/lf-6.6.36_2.1.0
- i.MX Smart Fitness: https://github.com/nxp-imx-support/imx-smart-fitness
- i.MX Smart Kitchen: https://github.com/nxp-imx-support/smart-kitchen
- i.MX E-Bike VIT: https://github.com/nxp-imx-support/imx-ebike-vit
