212 Hits in 2.7 sec

A Simple and efficient deep Scanpath Prediction [article]

Mohamed Amine Kerkouri, Aladine Chetouani
2021 arXiv   pre-print
Here, we explore the efficiency of using common deep learning architectures, in a simple fully convolutional regressive manner.  ...  Visual scanpath is the sequence of fixation points that the human gaze travels while observing an image, and its prediction helps in modeling the visual attention of an image.  ...  A Simple and efficient deep Scanpath Prediction Mohamed Amine KERKOURI, Aladine CHETOUANI  ... 
arXiv:2112.04610v1 fatcat:xc7qlk4odjc6fnjvg7dgsqxzbu

How Old Do You Look? Inferring Your Age from Your Gaze

A. Tianyi Zhang, B. Olivier Le Meur
2018 2018 25th IEEE International Conference on Image Processing (ICIP)  
Index Termsage inference, scanpath, deep network.  ...  In order to boost the performance, the training dataset is augmented by predicting a high number of scanpaths thanks to the use of an age-dependent computational saccadic model.  ...  CONCLUSION The proposed deep network predicts the age of an observer by scrutinizing his visual scanpath.  ... 
doi:10.1109/icip.2018.8451219 dblp:conf/icip/ZhangM18 fatcat:u5m6bttiz5ahxfi7bqwml44vna

COCO-Search18: A Dataset for Predicting Goal-directed Attention Control [article]

Yupei Chen, Zhibo Yang, Seoyoung Ahn, Dimitris Samaras, Minh Hoai, Gregory Zelinsky
2020 bioRxiv   pre-print
The currently best models of attention control are deep networks trained on free-viewing behavior to predict bottom-up attention control - saliency.  ...  model trained on behavioral search scanpaths.  ...  It is a search 471 efficiency metric because an initial saccade that lands directly 472 on the target would yield a Scanpath Ratio of 1, and all less 473 efficient searches would be < 1.  ... 
doi:10.1101/2020.07.27.221499 fatcat:kqa3x2wmtvfgpiiekxygxm3f6m

ST-MTL: Spatio-Temporal Multitask Learning Model to Predict Scanpath While Tracking Instruments in Robotic Surgery [article]

Mobarakol Islam, Vibashan VS, Chwee Ming Lim, Hongliang Ren
2021 arXiv   pre-print
We generate the task-aware saliency maps and scanpath of the instruments on the dataset of the MICCAI 2017 robotic instrument segmentation challenge.  ...  We also design a competitive squeeze and excitation unit by casting a skip connection that retains weak features, excites strong features, and performs dynamic spatial and channel-wise feature recalibration  ...  SalGAN [5] uses adversarial training over a deep CNN with a simple encoder-decoder architecture and binary cross-entropy (BCE) loss as an adversarial loss to estimate the saliency.  ... 
arXiv:2112.08189v1 fatcat:ck7rypz2lnfbplc4z3kejcjzae

Predicting Goal-directed Human Attention Using Inverse Reinforcement Learning [article]

Zhibo Yang, Lihan Huang, Yupei Chen, Zijun Wei, Seoyoung Ahn, Gregory Zelinsky, Dimitris Samaras, Minh Hoai
2020 arXiv   pre-print
When trained and evaluated on COCO-Search18, the IRL model outperformed baseline models in predicting search fixation scanpaths, both in terms of similarity to human search behavior and search efficiency  ...  These maps were learned by IRL and then used to predict behavioral scanpaths for multiple target categories.  ...  This project is supported by US National Science Foundation Award IIS-1763981, the Partner University Fund, the SUNY2020 Infrastructure Transportation Security Center, and a gift from Adobe.  ... 
arXiv:2005.14310v2 fatcat:svj53pymtnd5jj5v4jjau7ccl4

Deep Learning For Inter-Observer Congruency Prediction

Alexandre BRUCKERT, Yat Hong LAM, Marc CHRISTIE, Olivier LE MEUR
2019 2019 IEEE International Conference on Image Processing (ICIP)  
In this paper, we introduce a new method based on deep learning techniques to predict the IOC of an image.  ...  This is achieved by first extracting features from an image through a deep convolutional network.  ...  complex and efficient than simple linear regression.  ... 
doi:10.1109/icip.2019.8803596 dblp:conf/icip/BruckertLCM19 fatcat:3gz3z3l7wfhabinzrg42snwpxm

Generative Adversarial Reward Learning for Generalized Behavior Tendency Inference [article]

Xiaocong Chen, Lina Yao, Xianzhi Wang, Aixin Sun, Wenjie Zhang, Quan Z. Sheng
2021 arXiv   pre-print
traffic signal control, online recommender systems, and scanpath prediction.  ...  Our model provides a general way of characterizing and explaining underlying behavioral tendencies, and our experiments show our method outperforms state-of-the-art methods in a variety of scenarios, namely  ...  Scanpath Prediction Scanpath prediction is a type of goal-directed human intention prediction problem [15] . Take the last task in Fig. 1 for example.  ... 
arXiv:2105.00822v2 fatcat:itlyo4txfjbodp3wyb7oqehg3a

Gravitational Models Explain Shifts on Human Visual Attention [article]

Dario Zanca, Marco Gori, Stefano Melacci, Alessandra Rufa
2020 arXiv   pre-print
Another where the information from these maps is merged in order to select a single location to be attended for further and more complex computations and reasoning.  ...  Quantitative results on two large image datasets show that this model predicts shifts more accurately than winner-take-all.  ...  Acknowledgements We thank Frédéric Precioso and Lucile Sassatelli for fruitful discussions on the model which stimulated a wider view on the topic as well as interesting application perspectives in computer  ... 
arXiv:2009.06963v1 fatcat:37wvpihgxzewlefsi4u3udtsou

Scanpath modeling and classification with hidden Markov models

Antoine Coutrot, Janet H. Hsiao, Antoni B. Chan
2017 Behavior Research Methods  
However, eye movements are complex signals and many of these studies rely on limited gaze descriptors and bespoke datasets. Here, we provide a turnkey method for scanpath modeling and classification.  ...  Previous studies showed that scanpath, i.e., the sequence of eye movements made by an observer exploring a visual stimulus, can be used to infer observerrelated (e.g., task at hand) and stimuli-related  ...  We will focus on discriminant analysis as it includes both a predictive and a descriptive component: it is an efficient classification method, and it provides information on the relative importance of  ... 
doi:10.3758/s13428-017-0876-8 pmid:28409487 pmcid:PMC5809577 fatcat:quiexmw6rjbxpilppebb4eylm4

Relating Experience Goals With Visual User Interface Design

Jussi P P Jokinen, Johanna Silvennoinen, Tuomo Kujala
2018 Interacting with computers  
The authors present a cognitive top-down approach to this process, rooted in the appraisal theory and the theory of the predictive brain.  ...  The experience goals and repeated exposure to stimuli are shown to affect appraisal times and visual scanpaths in Web pages' evaluation; this supports the top-down approach described.  ...  This kind of predictive informationprocessing approach saves bandwidth and enables, for instance, efficient multitasking, ability to function in suboptimal conditions, and high plasticity for learning  ... 
doi:10.1093/iwc/iwy016 fatcat:i7kknqsdbfh7xm4qoqj4inodnu

Gravitational models explain shifts on human visual attention

Dario Zanca, Marco Gori, Stefano Melacci, Alessandra Rufa
2020 Scientific Reports  
Another where the information from these maps is merged in order to select a single location to be attended for further and more complex computations and reasoning.  ...  Quantitative results on two large image datasets show that this model predicts shifts more accurately than winner-take-all.  ...  Received: 23 February 2020; Accepted: 11 September 2020 Acknowledgements We thank Frédéric Precioso and Lucile Sassatelli for fruitful discussions on the model which stimulated a wider view on the topic  ... 
doi:10.1038/s41598-020-73494-2 pmid:33005008 fatcat:muug5twiorainm2xkfus3j2vqq

Predicting Goal-directed Attention Control Using Inverse-Reinforcement Learning [article]

Gregory J. Zelinsky, Yupei Chen, Seoyoung Ahn, Hossein Adeli, Zhibo Yang, Lihan Huang, Dimitrios Samaras, Minh Hoai
2020 arXiv   pre-print
We found that the IRL model predicted behavioral search efficiency and fixation-density maps using multiple metrics.  ...  Finally, we used these learned policies to predict the fixations of 60 new behavioral searchers (clock = 30, microwave = 30) in a disjoint test dataset of kitchen scenes depicting both a microwave and  ...  Acknowledgements We would like to thank the National Science Foundation for their generous support through award IIS-1763981, and members of the EyeCog Lab for their help with data collection and invaluable  ... 
arXiv:2001.11921v1 fatcat:zkrz3wbdqngtvab47dtu245dau

Processing Rhythmic Pattern during Chinese Sentence Reading: An Eye Movement Study

Yingyi Luo, Yunyan Duan, Xiaolin Zhou
2015 Frontiers in Psychology  
A V-O phrase could modify a noun by simply preceding it, forming a V-O-N compound; when the verb is disyllabic, however, the word order has to be O-V-N and the object is preferred to be disyllabic.  ...  Prosodic constraints play a fundamental role during both spoken sentence comprehension and silent reading.  ...  Gerrit Kentner and the two reviewers for their constructive comments and suggestions.  ... 
doi:10.3389/fpsyg.2015.01881 pmid:26696942 pmcid:PMC4673344 fatcat:5hcryhsjmfbuff46eiyx6whx2y

Image Quality Assessment without Reference by Combining Deep Learning-Based Features and Viewing Distance

Aladine Chetouani, Marius Pedersen
2021 Applied Sciences  
The results show the efficiency of our method and its generalization ability.  ...  For each patch, a feature vector is extracted from a convolutional neural network model and concatenated at the viewing distance, for which the quality is predicted.  ...  [46] extracted simple features from images by using a Shearlet transform, and then further treated image quality as a classification problem using deep neural networks.  ... 
doi:10.3390/app11104661 fatcat:rtkvff5e7ngxllvr4knr7ypim4

Perceptual Quality Assessment of Omnidirectional Images as Moving Camera Videos [article]

Xiangjie Sui, Kede Ma, Yiru Yao, Yuming Fang
2021 arXiv   pre-print
Moreover, we propose a computational framework for objective quality assessment of 360 images, embodying viewing conditions and behaviors in a delightful way.  ...  We construct a set of specific quality measures within the proposed framework, and demonstrate their promises on three VR quality databases.  ...  A simple and computationally efficient solution is to compute frame-level quality scores by IQA methods, followed by temporal pooling. Tu et al.  ... 
arXiv:2005.10547v2 fatcat:a2ccgnsydnblnd22lckgfjzuiy
« Previous Showing results 1 — 15 out of 212 results