Micro perceptual human computation for visual tasks

Gingold, Yotam; Shamir, Ariel; Cohen–Or, Daniel

doi:10.1145/2231816.2231817

Cited by 41 publications

(27 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Amazons Mechanical Turk is pioneering on-demand crowdsourcing that can draw on human computation or social computing systems. Research projects are exploring how this can be used for path planning [31], to determine depth layers, image normals, and symmetry from images [32], and to refine image segmentation [33]. Researchers 4 are working to understand pricing models [34] and apply crowdsourcing to grasping [35].…”

Section: E Crowdsourcing and Call Centersmentioning

confidence: 99%

IoT and Robotics: a synergy

Chowdhury¹

2017

Preprint

View full text Add to dashboard Cite

Abstract-The Internet of Robotic Things (IoRT) [9] is a concept first introduced by Dan Kara at ABI Research, which talks about augmenting the existing IoT with active sensorization; thereby, opening the doors to novel business ideas, at the intersection of both IoT and Robotics. This position paper considers the synergy between IoT and robotics: it talks about the technologies in IoT that would benefit the robotics domain. The advent of Cloud Robotics and its role in aiding robot functions like sensing, manipulation, and mobility. The paper then discusses the ways in which robots can extend the capabilities of existing IoT infrastructure by acting as a special class of edge device. IoT-aided robotic applications are discussed in various domains like health-care, military, industrial plants and rescue operations. The paper concludes by considering the use case of an Intelligent Transportation System endowed by an IoRT-inspired architecture.

show abstract

Section: E Crowdsourcing and Call Centersmentioning

confidence: 99%

IoT and Robotics: a synergy

Chowdhury¹

2017

Preprint

View full text Add to dashboard Cite

show abstract

“…Gingold et al [6] introduced a technique for applying the perception of a crowd to the analysis of 2D images. Chaudhuri et al [5] proposed AttribIt, an interface using semantic attributes for designing visual content, where they utilized crowds to learn the semantic attributes of design components.…”

Section: Crowd-powered Analysismentioning

confidence: 99%

“…Thus, we take a pairwise comparison approach [19,6,5] in which crowd workers are shown a pair of designs and asked to choose the best one. As a result, relative scores instead of absolute ones are obtained.…”

Section: Gathering Pairwise Comparisons By Crowdsourcingmentioning

confidence: 99%

Crowd-powered parameter analysis for visual design exploration

Koyama

Sakamoto

Igarashi

2014

Proceedings of the 27th Annual ACM Symposium on User Interface Software and Technology

View full text Add to dashboard Cite

Figure 1. Two interfaces for visual design exploration that are realized by our analysis technique. (Left) Smart Suggestion: The user can obtain appropriate parameter sets as suggestions. (Right) VisOpt Slider: The user can adjust each parameter effectively by the visualization (Vis) near the slider and the optimization (Opt) that gently guides the current parameters to the optimal direction. ABSTRACTParameter tweaking is one of the fundamental tasks in the editing of visual digital contents, such as correcting photo color or executing blendshape facial expression control. A problem with parameter tweaking is that it often requires much time and effort to explore a high-dimensional parameter space. We present a new technique to analyze such highdimensional parameter space to obtain a distribution of human preference. Our method uses crowdsourcing to gather pairwise comparisons between various parameter sets. As a result of analysis, the user obtains a goodness function that computes the goodness value of a given parameter set. This goodness function enables two interfaces for exploration: Smart Suggestion, which provides suggestions of preferable parameter sets, and VisOpt Slider, which interactively visualizes the distribution of goodness values on sliders and gently optimizes slider values while the user is editing. We created four applications with different design parameter spaces. As a result, the system could facilitate the user's design exploration.

show abstract

“…The use of crowdsourcing to collect data is gaining adoption, and has been used in recent approaches to a range of problems, including understanding shape through gauge figures [Cole et al 2009], creating a mesh segmentation database [Chen et al 2009], devising a retargeting evaluation framework [Rubinstein et al 2010], and for integrating humans into the loop for microtasks [Gingold et al 2012]. The experiences from this body of prior work has informed our design process.…”

Section: Related Workmentioning

confidence: 99%

“…In addition to providing positive feedback, we found it was necessary to prevent 159 users from doing our tasks because they were either malicious or had an accuracy below 50%. The use of sentinels is recommended by [Gingold et al 2012] to check labeler quality. However, we found that our tasks are not amenable to this approach since our surfaces occur in distinctive photographs, and repetition tips off the labeler.…”

Section: Task Analyticsmentioning

confidence: 99%

OpenSurfaces

et al. 2013

View full text Add to dashboard Cite

Consumer photographsSegmented surfaces Surface annotations scene: "kitchen" scene: "living room" di use, specular, roughness surface normal recti ed texture scene: "kitchen" object: "countertop" material: "granite" MaterialTexture Context Figure 1: We present OpenSurfaces, a large database of annotated surfaces created from real-world consumer photographs. Our annotation pipeline draws on crowdsourcing to segment surfaces from photos, and then annotates them with rich surface appearance properties, including material, texture, and contextual information. AbstractThe appearance of surfaces in real-world scenes is determined by the materials, textures, and context in which the surfaces appear. However, the datasets we have for visualizing and modeling rich surface appearance in context, in applications such as home remodeling, are quite limited. To help address this need, we present OpenSurfaces, a rich, labeled database consisting of thousands of examples of surfaces segmented from consumer photographs of interiors, and annotated with material parameters (reflectance, material names), texture information (surface normals, rectified textures), and contextual information (scene category, and object names).Retrieving usable surface information from uncalibrated Internet photo collections is challenging. We use human annotations and present a new methodology for segmenting and annotating materials in Internet photo collections suitable for crowdsourcing (e.g., through Amazon's Mechanical Turk). Because of the noise and variability inherent in Internet photos and novice annotators, designing this annotation engine was a key challenge; we present a multi-stage set of annotation tasks with quality checks and validation. We demonstrate the use of this database in proof-of-concept applications including surface retexturing and material and image browsing, and discuss future uses. OpenSurfaces is a public resource available at

show abstract

Micro perceptual human computation for visual tasks

Cited by 41 publications

References 54 publications

IoT and Robotics: a synergy

IoT and Robotics: a synergy

Crowd-powered parameter analysis for visual design exploration

OpenSurfaces

Contact Info

Product

Resources

About