Bookmark and Share

Technical Articles

The Vision AccelerationPac has specific hardware units and custom pipelines that are fully programmable from a high-level language.

Whether it is embedded processors, a full signal chain, interfaces or power, TI has the right solution for ADAS.

Many image processing applications in which humans will make the final interpretation are designed to optimize color reproduction for humans

Software is an integral part of, and is critical in building, ADAS systems that meets overall embedded product requirements.

This article provides an in-depth overview of VDMA use cases for TySOM reference designs.

The tools described in this paper help enable ADAS algorithms on automotive processors from Texas Instruments (TI).

For machine learning to deliver its potential, it requires sufficient high quality training data, plus developer knowledge of how to use it.


This article reviews deep learning implementation options such as heterogeneous processing, network quantization, and software optimizations


To achieve the lowest power with the best CNN graph performance in an ASIC or SoC, designers are turning to dedicated CNN engines.

The ability to stitch together multiple images, capturing varying viewpoints of a scene, is becoming increasingly valuable in applications.