VizWiz: nearly real-time answers to visual questions

By Jeffrey P. Bigham, Chandrika Jayant, H. Ji, Greg Little, Andrew Miller, Rob Miller, Aubrey Tatarowicz, Brandyn Allen White, Samuel White, Tom Yeh. 2010

247 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Networks and Communications, Computer Science Applications, Question Answering & Reading Comprehension

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Computer Networks and Communications

In Computer Science Applications

In Question Answering & Reading Comprehension

Citations in the corpus

VQA: Visual Question Answering (2015, neutral): “VQA [16, 30, 42, 2] is also amenable to automatic quantitative evaluation, making it possible to effectively track progress on this task.”
Connecting Vision and Language with Localized Narratives (2019, background): “One especially useful application for controlled captioning is assistive technology for people with imperfect vision [7,61,68], who could utilize the mouse to express their preferences in terms of how the image description should be presented.”
TextCaps: a Dataset for Image Captioning with Reading Comprehension (2020, background): “In concurrent work, [15] collect captions on VizWiz [5] images but unlike TextCaps there isn’t a specific focus on reading comprehension.”
Captioning Images Taken by People Who Are Blind (2020, methodology): “ds upon two existing datasets that contain images taken by real users of a visual description service [22,23]. The images in these datasets originate from users of the mobile phone application VizWiz [11], who each submitted a picture with, optionally, a recorded spoken question in order to receive ”
Leveraging Video Descriptions to Learn Video Question Answering (2016, methodology): “t self-paced learning is effective and the extended SS method outperforms other baselines. Related Work Image-QA. There has been a signiﬁcant recent interest in image-based visual question answering (Bigham et al. 2010; Geman et al. 2014; Malinowski and Fritz 2014; Malinowski, Rohrbach, and Fritz 20”
Privacy Concerns and Behaviors of People with Visual Impairments (2015, unknown): “One possibility is to use a combination of computer vision and crowdsourcing, which TapTapSee (reportedly) and Vizwiz [5] use to implement object detection on smartphones.”
Hands Holding Clues for Object Recognition in Teachable Machines (2019, methodology): “These results highlight the potential of our approach, given that photos of objects taken by people with visual impairments in the real world tend to include cluttered environments, as illustrated with the VizWiz dataset (Sec.”
Crowdsourcing subjective fashion advice using VizWiz: challenges and opportunities (2012, methodology): “As with the current instance of the VizWiz application [2], many of the responses and exchanges with users involved assisting them with taking better photos, an obvious caveat of the application.”
Pythia-A platform for vision & language research (2018, background): “0[10], Visual Dialog [6], VizWiz [3], COCO [14] and CLEVR [12] in Pythia.”
Answering visual questions with conversational crowd assistants (2013, methodology): “In contrast, recent crowd-powered systems such as VizWiz [2] have shown the ability of on-demand human computation to handle many common problems that blind users encounter.”
Crowdsourced Databases: Query Processing with People (2011, background): “present Quik Turkit [4], which implements several heuristics to assign HITs in near-realtime.”
Collaboratively crowdsourcing workflows with turkomatic (2012, background): “VizWiz is capable of handling open-ended, naturallanguage requests from users [7].”
Emotions Don't Lie: An Audio-Visual Deepfake Detection Method using Affective Cues (2020, background): “There are many other applications of video processing that use and combine multiple modalities for audio-visual speech recognition [28], emotion recognition [40, 63], and language and vision tasks [17, 29].”
Revolt: Collaborative Crowdsourcing for Labeling Machine Learning Datasets (2017, methodology): “Real-time crowdsourcing has been employed for a variety of purposes including minimizing reaction time in crowd-powered user interfaces [7, 6, 9, 34, 36], increasing the speed of data collection [35, 31], and improving data quality by incorporating expert interventions to guide novice workers [11,…”
Automatic Alt-text: Computer-generated Image Descriptions for Blind Users on a Social Network Service (2017, background): “While many existing systems caption images either with humans [1, 27, 34] or with computer algorithms [8, 9, 22], none of them fully satisfied the needs of visually impaired users on SNSs.”