Modeling Object Dissimilarity for Deep Saliency Prediction

TL;DR

This paper introduces a detection-guided deep saliency prediction network that explicitly models object dissimilarities, improving accuracy and outperforming state-of-the-art models on multiple benchmarks.

Contribution

It presents a novel method to incorporate object dissimilarity modeling into deep saliency networks, enhancing their predictive performance.

Findings

Outperforms state-of-the-art models on SALICON, MIT300, and CAT2000 benchmarks.

Explicit modeling of object dissimilarity boosts saliency prediction accuracy.

Fusion of dissimilarity features with existing networks improves baseline performance.

Abstract

Saliency prediction has made great strides over the past two decades, with current techniques modeling low-level information, such as color, intensity and size contrasts, and high-level ones, such as attention and gaze direction for entire objects. Despite this, these methods fail to account for the dissimilarity between objects, which affects human visual attention. In this paper, we introduce a detection-guided saliency prediction network that explicitly models the differences between multiple objects, such as their appearance and size dissimilarities. Our approach allows us to fuse our object dissimilarities with features extracted by any deep saliency prediction network. As evidenced by our experiments, this consistently boosts the accuracy of the baseline networks, enabling us to outperform the state-of-the-art models on three saliency benchmarks, namely SALICON, MIT300 and CAT2000. Our project page is at https://github.com/IVRL/DisSal.