A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf
.
Filters
Where To Look: Focus Regions for Visual Question Answering
[article]
2016
arXiv
pre-print
We present a method that learns to answer visual questions by selecting image regions relevant to the text-based query. ...
Our model is tested on the VQA dataset which is the largest human-annotated visual question answering dataset to our knowledge. ...
Conclusion We presented a model that learns to select regions from the image to solve visual question answering problems. ...
arXiv:1511.07394v2
fatcat:cjnltxfoqbfelmsadqbga623au
Where to Look: Focus Regions for Visual Question Answering
2016
2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
We present a method that learns to answer visual questions by selecting image regions relevant to the text-based query. ...
Our method maps textual queries and visual features from various regions into a shared space where they are compared for relevance with an inner product. ...
Conclusion We presented a model that learns to select regions from the image to solve visual question answering problems. ...
doi:10.1109/cvpr.2016.499
dblp:conf/cvpr/ShihSH16
fatcat:onvgqexleng6lg5ddddfs7xgvu
Intelligence, Where to Look, Where to Go?
2013
Journal of Intelligence
These are all important topics, some of which imply that more attention is given to other disciplines. For example, engineers are building intelligent systems and systems of artificial intelligence. ...
The suggestions for improvement are of a psychometric and substantive kind. ...
Attempting to answer such questions brings us back to the definitional issue.
Kind of Variation? ...
doi:10.3390/jintelligence1010005
fatcat:hprxc56po5dg3pdy5ylk3rfzxy
Knowing Where to Look? Analysis on Attention of Visual Question Answering System
[article]
2018
arXiv
pre-print
Attention mechanisms have been widely used in Visual Question Answering (VQA) solutions due to their capacity to model deep cross-domain interactions. ...
We find that both methods are sensitive to features, and simultaneously, they perform badly for counting and multi-object related questions. ...
For example in Fig. 1 , given a question about fire hydrant, we can see that MFB with object proposals focuses on the correct entity while imagelevel representation directs attentions to snow regions. ...
arXiv:1810.03821v1
fatcat:jt7tlr3linhf7miwx5ocsyyxre
Where to Look: A Unified Attention Model for Visual Recognition with Reinforcement Learning
[article]
2021
arXiv
pre-print
In this paper, we propose to unify the top-down and bottom-up attention together for recurrent visual attention. ...
Our model exploits the image pyramids and Q-learning to select regions of interests in the top-down attention mechanism, which in turn to guide the policy search in the bottom-up approach. ...
Anderson et al. (2018) proposes to combine bottom-up and top-down attention mechanism for image captioning and visual question answering. ...
arXiv:2111.07169v1
fatcat:l3nxgjoqffazrjvzgmcbu723e4
Knowing Where to Look? Analysis on Attention of Visual Question Answering System
[chapter]
2019
Lecture Notes in Computer Science
Attention mechanisms have been widely used in Visual Question Answering (VQA) solutions due to their capacity to model deep cross-domain interactions. ...
We ind that both methods are sensitive to features, and simultaneously, they perform badly for counting and multi-object related questions. ...
For example in Fig. 1 , given a question about ire hydrant, we can see that MFB with object proposals focuses on the correct entity while image-level representation directs attentions to snow regions. ...
doi:10.1007/978-3-030-11018-5_13
fatcat:hh5vzik7bnhnfk52pzcheufiai
Where-and-When to Look: Deep Siamese Attention Networks for Video-based Person Re-identification
[article]
2018
arXiv
pre-print
The network extracts local convolutional features from regions of each frame, and enhance their discriminative capability by focusing on distinct regions when measuring the similarity with another pedestrian ...
The model essentially learns which parts (where) from which frames (when) are relevant and distinctive for matching persons and attaches higher importance therein. ...
The idea is to directly comparing pairs of images and answer the question of whether the two images depict the same person or not. ...
arXiv:1808.01911v2
fatcat:hl64uomz4vfwpihbmelosrtkdi
Where to Look and How to Describe: Fashion Image Retrieval with an Attentional Heterogeneous Bilinear Network
[article]
2020
arXiv
pre-print
In order to distinguish images of different fashion products, we need to extract both appearance (i.e., "how to describe") and localization (i.e.,"where to look") information, and their interactions. ...
To this end, we propose a biologically inspired framework for image-based fashion product retrieval, which mimics the hypothesized twostream visual processing system of human brain. ...
[39] proposed a co-attention module for visual question answering that jointly performs visual attention and question attention. ...
arXiv:2010.13357v1
fatcat:bmwle4n5nbcq3fkvw6crxocdhi
Where You Look Matters for Body Perception: Preferred Gaze Location Contributes to the Body Inversion Effect
2017
PLoS ONE
The Body Inversion Effect (BIE; reduced visual discrimination performance for inverted compared to upright bodies) suggests that bodies are visually processed configurally; however, the specific importance ...
configuration processing to visual body discrimination. ...
Acknowledgments The authors would like to thank Galit Yovel for sharing stimuli and Paul Downing and Chris I. Baker for helpful remarks on an early draft of this manuscript. ...
doi:10.1371/journal.pone.0169148
pmid:28085894
pmcid:PMC5234795
fatcat:52utx665ynhivbjckuoogfzkhi
What/Where to Look Next? Modeling Top-Down Visual Attention in Complex Interactive Environments
2014
IEEE Transactions on Systems, Man & Cybernetics. Systems
Several visual attention models have been proposed for describing eye movements over simple stimuli and tasks such as free viewing or visual search. ...
In this study, we describe new task-dependent approaches for modeling top-down overt visual attention based on graphical models for probabilistic inference and reasoning. ...
Few attempts have been made to answer this question and existing models mainly apply to simple tasks such as visual search (e.g. [22] ). ...
doi:10.1109/tsmc.2013.2279715
fatcat:z4g6nari2jb5lbhx2jb2ouwyaq
Lesser-known or hidden reservoirs of infection and implications for adequate prevention strategies: Where to look and what to look for
2015
GMS Hygiene and Infection Control
In developing hygiene strategies, in recent years, the major focus has been on the hands as the key route of infection transmission. ...
In some instances, the causative organisms are particularly difficult to identify because they are concealed in biofilms or in a state referred to as viable but nonculturable, which eludes conventional ...
Many other questions have been raised which still need final answers, e.g., regarding the prerequisites for resuscitation, the infectivity of cells during the VBNC state, the effects in relation to disinfectants ...
doi:10.3205/dgkh000247
pmid:25699227
pmcid:PMC4332272
fatcat:b77reqq6rrcqfnrkemcx2qpa5i
Sexual dimorphism driven by intersexual resource competition: why is it rare, and where to look for it?
2021
Journal of Animal Ecology
Polygyny makes males disregard this female benefit, and both sexes compete for the most profitable resource, leading to overlapping niches. ...
Our models highlight that introducing conflict (achieved by switching from monogamy to polygamy) can also be responsible for sexual monomorphism. ...
Note that to aid visual comparison of male and female trait distributions at evolutionary equilibrium, our plots use a time axis that runs from left to right for males but from right to left for females ...
doi:10.1111/1365-2656.13487
pmid:33759459
fatcat:ccfku36swrfopip7ltpvfwrn2e
Acquisition and Use of 'Priors' in Autism: Typical in Deciding Where to Look, Atypical in Deciding What Is There
2020
Journal of Autism and Developmental Disorders
Individuals with ASD learned to avoid 'attentional capture' by distractors in the probable region as effectively as control participants—indicating typical priors for deploying attention. ...
This study examined whether this applies to decisions of attention allocation, of relevance for 'predictive-coding' accounts of ASD. ...
However, this scenario captures only part of how we deal with complex visual scenes, where a pre-attentive system of 'priority' computations determines where to allocate attention and post-selective perceptual ...
doi:10.1007/s10803-020-04828-2
pmid:33373014
pmcid:PMC8460564
fatcat:fovwryomkvc5jlrww2kxflc5jy
Looking for Creativity: Where Do We Look When We Look for New Ideas?
2016
Frontiers in Psychology
This 'looking at nothing' behavior has been observed during thinking that does not explicitly involve visual imagery (mind wandering, insight in problem solving, memory encoding and search) and it is associated ...
with reduced analysis of the external visual environment. ...
In fact, Ehrlichman and Weinberger (1978) found that participants were less likely to make eye movements (were more likely to stare) when answering visuospatial questions than when answering verbal questions ...
doi:10.3389/fpsyg.2016.00161
pmid:26913018
pmcid:PMC4753696
fatcat:ycuidnaxnbeodb7anq6sbfm5nm
Taking a new look at looking at nothing
2008
Trends in Cognitive Sciences
A crucial question in cognitive science is how linguistic and visual information are integrated. ...
Here, we discuss the looking at nothing phenomenon and use it to motivate a cognitive architecture for the integration of visual, spatial, linguistic and conceptual information. ...
For example, during reading, readers answering questions about previously read text will tend to move their eyes back to the location in the text containing the answer, even when the text is no longer ...
doi:10.1016/j.tics.2008.07.007
pmid:18805041
fatcat:qsutahbeircpln3s3hypha6ylm
« Previous
Showing results 1 — 15 out of 194,912 results