PILOT: A Pixel Intensity Driven Illuminant Color Estimation Framework for Color Constancy

06/25/2018
by   Jueqin Qiu, et al.
4

In this study, a CNN based Pixel Intensity driven iLluminant cOlor esTimation framework, PILOT, is proposed. The framework consists of a local illuminant estimation module and an illuminant uncertainty prediction module, obtained using a 3-phase training approach. The network with the well-designed microarchitecture of distillation building block and the macroarchitecture of bifurcated organization is of great representational capacity and efficacy for color-relevant vision tasks, which helps obtain a >20 over prior algorithms and achieve state-of-the-art accuracy of illuminant estimation on benchmark datasets. The proposed framework is also computationally efficient and parameter-economic, making it suitable for applications deployed on mobile platforms. The great interpretability also makes PILOT possible to serve as a guidance for designing statistics-based models for those low-end devices with tight budgets of power consumption and computational capacity.

READ FULL TEXT

page 1

page 7

page 9

research
01/28/2016

A Method for RFO Estimation Using Phase Analysis of Pilot Symbols in OFDM Systems

In this paper, a method for CFO/RFO estimation based on proportional coe...
research
12/19/2009

Secure Watermarking Scheme for Color Image Using Intensity of Pixel and LSB Substitution

In this paper a novel spatial domain LSB based watermarking scheme for c...
research
07/24/2023

A Good Student is Cooperative and Reliable: CNN-Transformer Collaborative Learning for Semantic Segmentation

In this paper, we strive to answer the question "how to collaboratively ...
research
04/19/2020

An end-to-end CNN framework for polarimetric vision tasks based on polarization-parameter-constructing network

Pixel-wise operations between polarimetric images are important for proc...
research
06/11/2019

Bag of Color Features For Color Constancy

In this paper, we propose a novel color constancy approach, called Bag o...
research
11/24/2021

Universal Deep Network for Steganalysis of Color Image based on Channel Representation

Up to now, most existing steganalytic methods are designed for grayscale...

Please sign up or login with your details

Forgot password? Click here to reset