Texas Instruments

Multi Camera AI on AM62A

This demo implements multiple camera AI application on AM62A processors based on the V3Link Camera Solution Kit and 4x IMX2a9 cameras. The demo shows how AM62A processed the videos and implemented object detection models at 120 FPS with bandwidth for expansion.

Texas Instruments

People Tracking on AM62A

This demo offers real-time vision-based people tracking with statistical insights such as total visitors, current occupancy, and visit duration. It also offers heatmap highlighting frequently visited areas. This demo has applications in areas such as retail, building automation, and security.

Texas Instruments

Monocular Depth Estimation

This demo shows single-camera depth estimation using a deep learning model. The MiDaS deep learning CNN gives relative depth information to distinguish people and objects from each other and backgrounds.

Texas Instruments

Image classification

Evaluate the performance of AI accelerators on TDA4x processors using different pre-compiled TensorFlow Lite, ONNX, or TVM models to classify images from USB and CSI camera inputs, as well as H.264 compressed video and JPEG image series.

Texas Instruments

Object detection

Evaluate the performance of AI accelerators on TDA4x processors using different pre-compiled TensorFlow Lite, ONNX, or TVM models to detect objects. Input sources include USB and CSI camera inputs, as well as H.264 compressed video and JPEG image series.

Texas Instruments

Semantic segmentation

Evaluate the performance of AI accelerators on TDA4x processors using different pre-compiled TensorFlow Lite, ONNX, or TVM semantic segmentation models. Input sources include USB and CSI camera inputs, as well as H.264 compressed video and JPEG image series.

Texas Instruments

Multiple input, multi-inference

This demo explains the data flow for two input sources passed through two inference operations—image classification, object detection and semantic segmentation. The outputs are overlaid on the input image individually and displayed using a mosaic layout. The single output image can be displayed on a screen or saved to H.264 compressed video or a JPG image series.

Texas Instruments

6D pose estimation

Object pose estimation aims to estimate the 3D orientation and 3D translation of objects in a given environment. It is useful in a wide range of applications like robotic manipulation for bin-picking, motion planning, and human-robot interaction task such as learning from demonstration. Supported on TDA4VM, AM68A and AM69A.

Texas Instruments

Human pose estimation

Multi person 2D human pose estimation is the task of understanding humans in an image. Given an input image, target is to detect each person and localize their body joints. Multi person 2D pose estimation is the task of understanding humans in an image. Given an input image, target is to detect each person and localize their body joints. Supported on TDA4VM, AM68A and AM69A.

Texas Instruments

Multi Camera AI on AM62A

This demo implements multiple camera AI application on AM62A processors based on the V3Link Camera Solution Kit and 4x IMX2a9 cameras. The demo shows how AM62A processed the videos and implemented object detection models at 120 FPS with bandwidth for expansion.

Texas Instruments

People Tracking on AM62A

This demo offers real-time vision-based people tracking with statistical insights such as total visitors, current occupancy, and visit duration. It also offers heatmap highlighting frequently visited areas. This demo has applications in areas such as retail, building automation, and security.

Texas Instruments

License plate recognition

License plate recognition aims to recognize the license plate number characters. This demo is specifically useful in wide area of applications including automotive domain. This demo is built on yolox-s-lite model, which is part of model zoo. Supported on TDA4VM.

Texas Instruments

Smart retail scanner

This demo showcases the use of vision AI on the AM62A processor for codeless food scanning, enabling automatic recognition of products and checkout of the user’s order. The demo employs an IMX219 camera and utilizes the built-in ISP and DL engine to perform real-time object detection inference.

Texas Instruments

Barcode imager with deep learning

This demo shows a barcode reader application for 1-D and 2-D codes. Using deep learning and open source decoder software, AM62A can be the main processor for high performance barcode readers and imagers.

Texas Instruments

Defect detection

Defect detection in machine vision and factory automation finds incorrectly produced parts and materials so they can be removed quickly and automatically. Edge AI Studio simplifies model training and deployment for new types of parts and materials.

Texas Instruments

Conferencing Systems

This demo showcases the use of vision AI and audio keyword spotting on AM6xA devices in the context of audio-visual conferencing systems. Audio commands control the area that the camera focuses on.

Texas Instruments

Gesture controlled HMI

This design shows integration of mmWave radar and camera sensor with AM62/AM62A to control a building access HMI with face detection and gestures to unlock a PIN controlled entry

Texas Instruments

Smart camera with Vision-Radar Fusion

Vision and Radar fusion covers weakness in the individual sensors like poor lighting or low resolution. Radar sensors detect movement in 3D space, allowing better proximity detection while vision on AM62/AM62A recognizes objects like vehicles, people or animals.

Texas Instruments

Stereo depth estimation

This Robot Operating System (ROS) application demonstrates hardware accelerated stereo vision processing on a live stereo camera or a ROS bag file on TDA4VM Processor. Computation intensive tasks such as image rectification, scaling and stereo disparity estimation are processed on vision hardware accelerators.

Texas Instruments

Environmental awareness

This Robot Operating System (ROS) application demonstrates a hardware accelerated semantic segmentation task on live camera or ROS bag data using the TDA4VM processor. Computation intensive image pre-processing tasks such as image rectification and scaling happen on vision hardware accelerator VPAC, while the AI processing is accelerated on deep learning accelerator.

Texas Instruments

3D obstacle detection

This ROS based application demonstrates hardware accelerated 3D obstacle detection on a stereo vision input. The hybrid computer vision and AI based demo utilizes vision hardware accelerators on TDA4VM for image rectification, stereo disparity mapping and scaling tasks, while AI-based semantic segmentation is accelerated on deep learning accelerator.

Texas Instruments

Visual localization

This application demonstrates hardware accelerated ego vehicle localization, estimating 6-degrees of freedom pose. It uses a deep neural network to learn hand computed feature descriptor like KAZE in a supervised manner. The AI and CV based demo utilizes deep learning hardware accelerator and TI DSP on TDA4VM to run the task efficiently, while leaving the Arm® cores completely free for other tasks.

Texas Instruments

Monocular Depth Estimation

This demo shows single-camera depth estimation using a deep learning model. The MiDaS deep learning CNN gives relative depth information to distinguish people and objects from each other and backgrounds.