Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

FastSal: a Computationally Efficient Network for Visual Saliency Prediction

About

This paper focuses on the problem of visual saliency prediction, predicting regions of an image that tend to attract human visual attention, under a constrained computational budget. We modify and test various recent efficient convolutional neural network architectures like EfficientNet and MobileNetV2 and compare them with existing state-of-the-art saliency models such as SalGAN and DeepGaze II both in terms of standard accuracy metrics like AUC and NSS, and in terms of the computational complexity and model size. We find that MobileNetV2 makes an excellent backbone for a visual saliency model and can be effective even without a complex decoder. We also show that knowledge transfer from a more computationally expensive model like DeepGaze II can be achieved via pseudo-labelling an unlabelled dataset, and that this approach gives result on-par with many state-of-the-art algorithms with a fraction of the computational cost and model size. Source code is available at https://github.com/feiyanhu/FastSal.

Feiyan Hu, Kevin McGuinness• 2020

Related benchmarks

TaskDatasetResultRank
Saliency PredictionSALICON (test)
NSS1.845
25
Visual Saliency PredictionCAT2000 (test)
Correlation Coefficient (CC)0.7213
19
Saliency PredictionMIT1003 (test)
NSS2.0078
18
Saliency PredictionCAT2000 Natural scene
CC0.721
8
Saliency PredictionMIT1003 Natural scene
CC0.59
6
Showing 5 of 5 rows

Other info

Follow for update