Design Tools for Application Specific Embedded Processors

On Object-Oriented Databases, Materialized Views, and Concurrent Engineering

ASME 1991 5th Annual Database Symposium: Engineering Databases — An Enterprise Resource ◽

10.1115/edm1991-0187 ◽

1991 ◽

Author(s):

Martin Hardwick ◽

Blair R. Downie

Keyword(s):

Concurrent Engineering ◽

Object Oriented ◽

Materialized Views ◽

Design Tools ◽

Design Data ◽

Multiple Inheritance ◽

File Formats ◽

Definition Of ◽

Application Specific ◽

Object Oriented Database

Abstract Concurrent engineering seeks to reduce the length of the design life cycle by allowing multiple engineers to work on a design concurrently using their different design tools. A major stumbling block in achieving this goal is that most design tools use different file formats. Emerging standards such as STEP/PDES/EXPRESS reduce this barrier, but conformance to standards is not enough. One reason design tools have different file formats is because each tool requires a different perspective or view of the design. Engineering databases must provide designers with the ability to define application specific views of design data, and the ability to propagate changes among those related views. In this paper, we examine how an object-oriented database system can support the definition of application views using a class hierarchy and multiple inheritance.

Download Full-text

A power reduction technique with object code merging for application specific embedded processors

Proceedings Design, Automation and Test in Europe Conference and Exhibition 2000 (Cat. No. PR00537) ◽

10.1109/date.2000.840849 ◽

2002 ◽

Cited By ~ 6

Author(s):

T. Ishihara ◽

H. Yasuura

Keyword(s):

Power Reduction ◽

Reduction Technique ◽

Embedded Processors ◽

Object Code ◽

Application Specific

Download Full-text

Differentiable Compound Optics and Processing Pipeline Optimization for End-to-end Camera Design

ACM Transactions on Graphics ◽

10.1145/3446791 ◽

2021 ◽

Vol 40 (2) ◽

pp. 1-19

Author(s):

Ethan Tseng ◽

Ali Mosleh ◽

Fahim Mannan ◽

Karl St-Arnaud ◽

Avinash Sharma ◽

...

Keyword(s):

Neural Network ◽

Image Processing ◽

Optical Design ◽

Neural Nets ◽

Fine Tuning ◽

Optical Systems ◽

Design Tools ◽

End To End ◽

Application Specific ◽

Network Processing

Most modern commodity imaging systems we use directly for photography—or indirectly rely on for downstream applications—employ optical systems of multiple lenses that must balance deviations from perfect optics, manufacturing constraints, tolerances, cost, and footprint. Although optical designs often have complex interactions with downstream image processing or analysis tasks, today’s compound optics are designed in isolation from these interactions. Existing optical design tools aim to minimize optical aberrations, such as deviations from Gauss’ linear model of optics, instead of application-specific losses, precluding joint optimization with hardware image signal processing (ISP) and highly parameterized neural network processing. In this article, we propose an optimization method for compound optics that lifts these limitations. We optimize entire lens systems jointly with hardware and software image processing pipelines, downstream neural network processing, and application-specific end-to-end losses. To this end, we propose a learned, differentiable forward model for compound optics and an alternating proximal optimization method that handles function compositions with highly varying parameter dimensions for optics, hardware ISP, and neural nets. Our method integrates seamlessly atop existing optical design tools, such as Zemax . We can thus assess our method across many camera system designs and end-to-end applications. We validate our approach in an automotive camera optics setting—together with hardware ISP post processing and detection—outperforming classical optics designs for automotive object detection and traffic light state detection. For human viewing tasks, we optimize optics and processing pipelines for dynamic outdoor scenarios and dynamic low-light imaging. We outperform existing compartmentalized design or fine-tuning methods qualitatively and quantitatively, across all domain-specific applications tested.

Download Full-text