DeepAI AI Chat
Log In Sign Up

Influence of Image Classification Accuracy on Saliency Map Estimation

07/27/2018
by   Taiki Oyama, et al.
Sophia School Corp.
0

Saliency map estimation in computer vision aims to estimate the locations where people gaze in images. Since people tend to look at objects in images, the parameters of the model pretrained on ImageNet for image classification are useful for the saliency map estimation. However, there is no research on the relationship between the image classification accuracy and the performance of the saliency map estimation. In this paper, it is shown that there is a strong correlation between image classification accuracy and saliency map estimation accuracy. We also investigated the effective architecture based on multi scale images and the upsampling layers to refine the saliency-map resolution. Our model achieved the state-of-the-art accuracy on the PASCAL-S, OSIE, and MIT1003 datasets. In the MIT Saliency Benchmark, our model achieved the best performance in some metrics and competitive results in the other metrics.

READ FULL TEXT

page 1

page 3

page 8

page 11

07/24/2020

Hallucinating Saliency Maps for Fine-Grained Image Classification for Limited Data Domains

Most of the saliency methods are evaluated on their ability to generate ...
07/17/2018

Saliency Map Estimation for Omni-Directional Image Considering Prior Distributions

In recent years, the deep learning techniques have been applied to the e...
10/14/2017

Co-saliency Detection for RGBD Images Based on Multi-constraint Feature Matching and Cross Label Propagation

Co-saliency detection aims at extracting the common salient regions from...
10/06/2020

Visualizing Color-wise Saliency of Black-Box Image Classification Models

Image classification based on machine learning is being commonly used. H...
08/23/2018

Maximal Jacobian-based Saliency Map Attack

The Jacobian-based Saliency Map Attack is a family of adversarial attack...
09/05/2021

Deep Saliency Prior for Reducing Visual Distraction

Using only a model that was trained to predict where people look at imag...
02/01/2019

ColorNet: Investigating the importance of color spaces for image classification

Image classification is a fundamental application in computer vision. Re...