ImageSpirit
Humans describe images in terms of nouns and adjectives while algorithms operate on images represented as sets of pixels. Bridging this gap between how humans would like to access images versus their typical representation is the goal of image parsing, which involves assigning object and attribute l...
Main Authors: | , , , , , , , |
---|---|
Format: | Journal article |
Sprog: | English |
Udgivet: |
Association for Computing Machinery
2014
|
_version_ | 1826313275685470208 |
---|---|
author | Cheng, M-M Zheng, S Lin, W-Y Vineet, V Sturgess, P Crook, N Mitra, NJ Torr, P |
author_facet | Cheng, M-M Zheng, S Lin, W-Y Vineet, V Sturgess, P Crook, N Mitra, NJ Torr, P |
author_sort | Cheng, M-M |
collection | OXFORD |
description | Humans describe images in terms of nouns and adjectives while algorithms operate on images represented as sets of pixels. Bridging this gap between how humans would like to access images versus their typical representation is the goal of image parsing, which involves assigning object and attribute labels to pixels. In this article we propose treating nouns as object labels and adjectives as visual attribute labels. This allows us to formulate the image parsing problem as one of jointly estimating per-pixel object and attribute labels from a set of training images. We propose an efficient (interactive time) solution. Using the extracted labels as handles, our system empowers a user to verbally refine the results. This enables hands-free parsing of an image into pixel-wise object/attribute labels that correspond to human semantics. Verbally selecting objects of interest enables a novel and natural interaction modality that can possibly be used to interact with new generation devices (e.g., smartphones, Google Glass, livingroom devices). We demonstrate our system on a large number of real-world images with varying complexity. To help understand the trade-offs compared to traditional mouse-based interactions, results are reported for both a large-scale quantitative evaluation and a user study. |
first_indexed | 2024-09-25T04:10:30Z |
format | Journal article |
id | oxford-uuid:66b8c923-aa5f-4cfc-952f-b6637d88d87c |
institution | University of Oxford |
language | English |
last_indexed | 2024-09-25T04:10:30Z |
publishDate | 2014 |
publisher | Association for Computing Machinery |
record_format | dspace |
spelling | oxford-uuid:66b8c923-aa5f-4cfc-952f-b6637d88d87c2024-06-24T15:17:29ZImageSpiritJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:66b8c923-aa5f-4cfc-952f-b6637d88d87cEnglishSymplectic ElementsAssociation for Computing Machinery2014Cheng, M-MZheng, SLin, W-YVineet, VSturgess, PCrook, NMitra, NJTorr, PHumans describe images in terms of nouns and adjectives while algorithms operate on images represented as sets of pixels. Bridging this gap between how humans would like to access images versus their typical representation is the goal of image parsing, which involves assigning object and attribute labels to pixels. In this article we propose treating nouns as object labels and adjectives as visual attribute labels. This allows us to formulate the image parsing problem as one of jointly estimating per-pixel object and attribute labels from a set of training images. We propose an efficient (interactive time) solution. Using the extracted labels as handles, our system empowers a user to verbally refine the results. This enables hands-free parsing of an image into pixel-wise object/attribute labels that correspond to human semantics. Verbally selecting objects of interest enables a novel and natural interaction modality that can possibly be used to interact with new generation devices (e.g., smartphones, Google Glass, livingroom devices). We demonstrate our system on a large number of real-world images with varying complexity. To help understand the trade-offs compared to traditional mouse-based interactions, results are reported for both a large-scale quantitative evaluation and a user study. |
spellingShingle | Cheng, M-M Zheng, S Lin, W-Y Vineet, V Sturgess, P Crook, N Mitra, NJ Torr, P ImageSpirit |
title | ImageSpirit |
title_full | ImageSpirit |
title_fullStr | ImageSpirit |
title_full_unstemmed | ImageSpirit |
title_short | ImageSpirit |
title_sort | imagespirit |
work_keys_str_mv | AT chengmm imagespirit AT zhengs imagespirit AT linwy imagespirit AT vineetv imagespirit AT sturgessp imagespirit AT crookn imagespirit AT mitranj imagespirit AT torrp imagespirit |