ABSTRACT
Existing studies used to estimate visual attention with the bottom-up (e.g., image salient features) and top-down methods (e.g., task-driven information searching patterns) and have gained progressive results in computing probable visual attention with various image types, user groups, and viewing durations. However, these works are mostly based on fixed distances of viewing and cannot be generalised to visual scanpath estimation at dynamic viewing distances. Therefore, the research fills this gap by 1) investigating user's visual attention patterns at different viewing distances, and 2) developing the distance-based visual scanpath estimation model that can generate human-like visual scanpath at different distances. The research also prepares a large-scale eye tracking dataset to support the visual scanpath model and evaluates it with applications such as graphic design and advertisement. The research's main contributions are two-fold. Firstly, it provides the novel visual scanpath estimation model that works at different viewing distances and secondly, draws insights into how effective of the model in different applications.
- Fosco, C., , Predicting Visual Importance Across Graphic Design Types, in UIST'20. 2020, Association for Computing Machinery: Virtual Event, USA. p. 249–260.Google ScholarDigital Library
- Meur, O.L. and A. Coutrot, How saccadic models help predict where we look during a visual task? Application to visual quality assessment. Electronic Imaging, 2016. 2016(13): p. 1-7.Google Scholar
- Yun, K., , Exploring the role of gaze behavior and object detection in scene understanding. Frontiers in psychology, 2013. 4: p. 917.Google Scholar
- Fosco, C., How much time do you have? modeling multi-duration saliency. in CVPR’20. 2020.Google ScholarCross Ref
- Borji, A. and L. Itti, Cat2000: A large scale fixation dataset for boosting saliency research. arXiv preprint arXiv:1505.03581, 2015.Google Scholar
- Judd, T., Learning to predict where humans look. in 2009 IEEE 12th international conference on computer vision. 2009. IEEE.Google ScholarCross Ref
- Sun, W., Z. Chen, and F. Wu, Visual Scanpath Prediction Using IOR-ROI Recurrent Mixture Density Network. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021. 43(6): p. 2101-2118.Google ScholarCross Ref
- Todd, J.T. and J.F. Norman, The visual perception of 3-D shape from multiple cues: Are observers capable of perceiving metric structure? Perception & Psychophysics, 2003. 65(1): p. 31-47.Google ScholarCross Ref
- Qian, J. and Y. Petrov, A depth illusion supports the model of General Object Constancy: Size and depth constancies related by a same distance-scaling factor. Vision Research, 2016. 129: p. 77-86.Google ScholarCross Ref
- Carrasco, M., P.E. Williams, and Y. Yeshurun, Covert attention increases spatial resolution with or without masks: Support for signal enhancement. Journal of vision, 2002. 2(6): p. 4-4.Google ScholarCross Ref
- Carrasco, M., C.P. Talgar, and E.L. Cameron, Characterizing visual performance fields: Effects of transient covert attention, spatial frequency, eccentricity, task and set size. Spatial vision, 2001. 15(1): p. 61-75.Google ScholarCross Ref
- McCourt, M.E. and M. Garlinghouse, Asymmetries of visuospatial attention are modulated by viewing distance and visual field elevation: Pseudoneglect in peripersonal and extrapersonal space. Cortex, 2000. 36(5): p. 715-731.Google ScholarCross Ref
- Tong, S., Visual attention inspired distant view and close-up view classification. in 2016 IEEE International Conference on Image Processing (ICIP). 2016. IEEE.Google ScholarCross Ref
- Itti, L., C. Koch, and E. Niebur, A model of saliency-based visual attention for rapid scene analysis. IEEE Transactions on pattern analysis and machine intelligence, 1998. 20(11): p. 1254-1259.Google ScholarDigital Library
- Le Meur, O. and Z. Liu, Saccadic model of eye movements for free-viewing condition. Vision research, 2015. 116: p. 152-164.Google Scholar
- Wang, W., Simulating human saccadic scanpaths on natural images. in CVPR 2011. 2011. IEEE.Google ScholarDigital Library
- Wloka, C., I. Kotseruba, and J.K. Tsotsos. Active fixation control to predict saccade sequences. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018.Google ScholarCross Ref
- Jiang, M., , Learning to predict sequences of human visual fixations. IEEE transactions on neural networks and learning systems, 2016. 27(6): p. 1241-1252.Google ScholarCross Ref
- Malem-Shinitski, N., , A mathematical model of local and global attention in natural scene viewing. PLoS Computational Biology, 2020. 16(12): p. e1007880.Google ScholarCross Ref
- Assens, M., , Scanpath and saliency prediction on 360 degree images. Signal Processing: Image Communication, 2018. 69: p. 8-14.Google ScholarCross Ref
- Assens, M., PathGAN: Visual scanpath prediction with generative adversarial networks. in Proceedings of the European Conference on Computer Vision (ECCV) Workshops. 2018.Google Scholar
- Kümmerer, M., M. Bethge, and T.S. Wallis, DeepGaze III: Modeling free-viewing human scanpaths with deep learning. Journal of Vision, 2022. 22(5): p. 7-7.Google ScholarCross Ref
- Kümmerer, M., T. Wallis, and M. Bethge, Deepgaze ii: Predicting fixations from deep features over time and tasks. Journal of Vision, 2017. 17(10): p. 1147-1147.Google ScholarCross Ref
- Alhashim, I. and P. Wonka, High quality monocular depth estimation via transfer learning. arXiv preprint arXiv:1812.11941, 2018.Google Scholar
- Cristino, F., , ScanMatch: A novel method for comparing fixation sequences. Behavior research methods, 2010. 42(3): p. 692-700.Google Scholar
- Dewhurst, R., , It depends on how you look at it: Scanpath comparison in multiple dimensions with MultiMatch, a vector-based approach. Behavior research methods, 2012. 44(4): p. 1079-1100.Google Scholar
Index Terms
- Distance-based Visual Scanpath Estimation and Applications
Recommendations
Personalized Visual Scanpath Prediction Using IOR-ROI Weighted Attention Network
IMXw '23: Proceedings of the 2023 ACM International Conference on Interactive Media Experiences WorkshopsPredicting visual scanpaths plays an important role in modeling overt human visual attention and search behavior. Due to the rapid development of deep learning, previous scanpath prediction models have made significant progress. However, these methods ...
Human Visual Scanpath Prediction Based on RGB-D Saliency
ICIGP '18: Proceedings of the 2018 International Conference on Image and Graphics ProcessingHuman visual perception is considered as a dynamic process of information acquisition, while the visual scanpath can clearly reflect the shift of our eye fixations. In the previous study of visual attention, researchers generally do the saliency ...
Target Search by Bottom-Up and Top-Down Fuzzy Information
SIBGRAPI '00: Proceedings of the 13th Brazilian Symposium on Computer Graphics and Image ProcessingOne of the basic tasks assigned to the attentional mechanism is to decide which location in the visual field we must pay attention first. An object containing a distinctive feature can attract attention in a bottom-up way. By comparing one object with ...
Comments