Process This: Efficient stereo depth estimation and 3D perception for spatial AI applications
Our world is three dimensional. So, it’s critical that we—and the applications that we build—have depth perception and can see in 3D. Whether you are building an autonomous robot, vehicle or machine, or creating a machine vision camera with embedded AI, 3D computer vision technology is critical.
In this webinar, we focus on Stereo depth estimation (SDE), a technique that uses two cameras and epipolar geometry to calculate the depth of different objects in a scene. This technique requires high levels of computing resources. We explain how TI’s TDA4VM edge AI processor, with high levels of integration, has a special hardware accelerator for SDE which is optimized for high-performance, real-time computer vision algorithms such as localization, mapping, scene analysis and object detection in automotive and Robotics.
- Stereo depth estimation (SDE) concepts
- Use cases in industrial and automotive applications
- Efficient SDE using hardware accelerators
- Performance optimized image rectification
- Live demonstration of the software framework and application examples