A domain adaptive deep learning solution for scanpath prediction of paintings

Abstract

Cultural heritage understanding and preservation is an important issue for society as it represents a fundamental aspect of its identity. Paintings represent a significant part of cultural heritage, and are the subject of study continuously. However, the way viewers perceive paintings is strictly related to the so-called HVS (Human Vision System) behaviour. This paper focuses on the eye-movement analysis of viewers during the visual experience of a certain number of paintings. In further details, we introduce a new approach to predicting human visual attention, which impacts several cognitive functions for humans, including the fundamental understanding of a scene, and then extend it to painting images. The proposed new architecture ingests images and returns scanpaths, a sequence of points featuring a high likelihood of catching viewers' attention. We use an FCNN (Fully Convolutional Neural Network), in which we exploit a differentiable channel-wise selection and Soft-Argmax modules. We also incorporate learnable Gaussian distributions onto the network bottleneck to simulate visual attention process bias in natural scene images. Furthermore, to reduce the effect of shifts between different domains (i.e. natural images, painting), we urge the model to learn unsupervised general features from other domains using a gradient reversal classifier. The results obtained by our model outperform existing state-of-the-art ones in terms of accuracy and efficiency.

A review of machine learning in scanpath analysis for passive gaze-based interaction

“…The studies by Bao and Chen ( 2020 ), Kerkouri et al ( 2022 ), and Kümmerer et al ( 2022 ) all used the SALICON dataset (Jiang et al, 2015 ) to train their networks. Bao and Chen ( 2020 ) and Kerkouri et al ( 2022 ) trained a ResNet-50, and a MobileNet (Sandler et al, 2018 ) with an additional CNN and MLP networks, respectively, to predict visual attention based on saliency maps. Kümmerer et al ( 2022 ) trained a DenseNet201 (Huang et al, 2017 ) with two additional 2-layer convolutional networks to predict upcoming fixations from prior ones.…”

“…Semmelrock et al ( 2023 ) attributed this to multiple reasons, such as different data, package versions, hardware setups, and non-determinism of ML models, which is why using fixed random seeds is vital. For example, the code link attached to this publication (Kerkouri et al, 2022 ) states the code will be made available, and has not been updated ever since.…”

Data_Sheet_1.xlsx

“…The studies by Bao and Chen (2020), Kerkouri et al (2022), and Kümmerer et al (2022) all used the SALICON dataset (Jiang et al, 2015) to train their networks. Bao and Chen (2020) and Kerkouri et al (2022) trained a ResNet-50, and a MobileNet (Sandler et al, 2018) with an additional CNN and MLP networks, respectively, to predict visual attention based on saliency maps. Kümmerer et al (2022) trained a DenseNet201 (Huang et al, 2017) with two additional 2-layer convolutional networks to predict upcoming fixations from prior ones.…”

“…Semmelrock et al (2023) attributed this to multiple reasons, such as different data, package versions, hardware setups, and nondeterminism of ML models, which is why using fixed random seeds is vital. For example, the code link attached to this publication (Kerkouri et al, 2022) states the code will be made available, and has not been updated ever since.…”