Brightening the Optical Flow through Posit Arithmetic

@article{Saxena2021BrighteningTO,
  title={Brightening the Optical Flow through Posit Arithmetic},
  author={V. Saxena and A. Reddy and J. Neudorfer and J. Gustafson and Sangeeth Nambiar and R. Leupers and Farhad Merchant},
  journal={2021 22nd International Symposium on Quality Electronic Design (ISQED)},
  year={2021},
  pages={463-468}
}
As new technologies are invented, their commercial viability needs to be carefully examined along with their technical merits and demerits. The ${posit}^{TM}$ data format, proposed as a drop-in replacement for IEEE $754 ^{TM}$ float format, is one such invention that requires extensive theoretical and experimental study to identify products that can benefit from the advantages of posits for specific market segments. In this paper, we present an extensive empirical study of posit-based… Expand

Figures and Tables from this paper

An Investigation on Inherent Robustness of Posit Data Representation
TLDR
A comparative robustness study between 32-bit posit and 32- bit IEEE 754–2008 compliant representations is proposed and found that posit representation is less impacted by faults than the classical floating-point-based representation. Expand

References

SHOWING 1-10 OF 25 REFERENCES
PERC: Posit Enhanced Rocket Chip
Balancing between precision and performance is a known trade-off in system design. Universal Number system attempts to dissolve this trade-off with its flexible arbitrary precision bound within fixedExpand
Evaluating the Hardware Cost of the Posit Number System
TLDR
This article introduces an open-source hardware implementation of the posit number system, in the form of a C++ templatized library compatible with Vivado HLS, and includes the first open- source parameterized hardware quire. Expand
Cheetah: Mixed Low-Precision Hardware & Software Co-Design Framework for DNNs on the Edge
TLDR
The Cheetah framework is proposed, which supports both DNN training and inference using posits, as well as other commonly used formats and is amenable for different quantization approaches and supports mixed-precision floating point and fixed-point numerical formats. Expand
PERI: A Posit Enabled RISC-V Core
TLDR
The paper provides insights on how the current 'F' extension and the custom op-code space of RISC-V can be leveraged/modified to support Posit arithmetic and presents a workaround on how certain applications can be modified minimally to exploit the existing Risc-V tool-chain. Expand
A Novel Hardware Architecture of the Lucas–Kanade Optical Flow for Reduced Frame Memory Access
TLDR
A novel hardware architecture is proposed that stores the input image after the Gaussian filtering operation instead of the original input image itself, which reduces the frame memory access by 61%-75% compared with the previous research. Expand
An Investigation on Inherent Robustness of Posit Data Representation
TLDR
A comparative robustness study between 32-bit posit and 32- bit IEEE 754–2008 compliant representations is proposed and found that posit representation is less impacted by faults than the classical floating-point-based representation. Expand
ExPAN(N)D: Exploring Posits for Efficient Artificial Neural Network Design in FPGA-Based Systems
TLDR
This paper presents ExPAN(N)D, a framework to analyze and ingather the efficacy of the Posit number representation scheme and the efficiency of fixed-point arithmetic implementations for ANNs, and proposes a modified Posit-based representation to store the trained parameters of a network. Expand
Parameterized Posit Arithmetic Hardware Generator
TLDR
The architecture of a parameterized PAU generator that can generate PAU adders and PAU multipliers of any bit-width pre-synthesis is presented and it is argued that an n-bit IEEE 754-2008 adder and multiplier can be safely replaced with an m-bit PAU addition and multiplier where m<n, due to superior numerical accuracy of the PAU. Expand
PACoGen: A Hardware Posit Arithmetic Core Generator
TLDR
This paper proposes open-source hardware Posit Arithmetic Core Generator (PACoGen) for the recently developed universal number posit number system, along with a set of pipelined architectures for addition/subtraction, multiplication, and division arithmetic. Expand
Beating Floating Point at its Own Game: Posit Arithmetic
TLDR
A new data type called a posit is designed as a direct drop-in replacement for IEEE Standard 754 floating-point numbers (floats), and provides compelling advantages over floats, including larger dynamic range, higher accuracy, better closure, bitwise identical results across systems, simpler hardware, and simpler exception handling. Expand
...
1
2
3
...