Object detection, tracking and template matching.

OVERVIEW


beyond computer vision

APPLICATIONS


accelerating the future

OVERVIEW

blue_small_bar

beyond computer vision

APPLICATIONS

blue_small_bar

accelerating the future

OVERVIEW

  • NAME: Object Detection Engine (ODE)
  • USAGE: object detection, tracking and template matching
  • HOW: performs intensive computational tasks with reduced power consumption, bandwidth and execution times
  • IP: ODE is a patented hardware engine
  • PART OF: Image Processing Unit (IPU)
  • TYPE: computer vision core, user experience technology
  • INTEGRATION: hardware
  • MARKETS:1  2  3  6  8  9   11  12   

The Object Detection Engine (ODE) is a state-of-the-art engine from FotoNation that addresses current problems and limitations in the image processing pipeline.

ODE is used for computationally intensive tasks, with control and configuration performed in software.

This unique implementation is based on pre-processing on-the-fly, as the images get transferred from the ISP to memory, with minimum R/W accesses to memory and by using small resolution images for global analysis and normalized windows for focal analysis.

Designed to work with minimum resources – very low processor and power usage, ODE accelerates and enables a wide range of features that can run in parallel.

APPLICATIONS

Computer vision algorithms are at the foundation of some of the most outstanding implementations of today’s embedded systems.

Understanding images has become key in deploying or improving existing use-cases across various industries. The continuous technological improvements have led to highly specialized systems that can detect and track an object with incredible accuracy.

The Object Filtering Engine is the first of its kind programmable engine that can be tailored to a specific scenario with virtually no limit of objects detected. It can adapt to any scenario or industry while delivering the best results, even in difficult illumination conditions or from systems with restrictions in terms of power consumption or bandwidth.

Fast and precise, with detection data available in every frame, our engine is the perfect choice to accelerate computer vision algorithms for face detection and tracking, face features detection, smile and blink detection, face recognition, face beautification or body detection and tracking.

Besides these straightforward applications, the engine can be programmed to accelerate the detection and tracking of generic objects with no-trade-offs and the same performance and precision, leading to never seen before use-cases.