DASH: Visual Analytics for Debiasing Image Classification via User-Driven Synthetic Data Augmentation

dc.contributor.authorKwon, Bum Chulen_US
dc.contributor.authorLee, Jungsooen_US
dc.contributor.authorChung, Chaeyeonen_US
dc.contributor.authorLee, Nyoungwooen_US
dc.contributor.authorChoi, Ho-Jinen_US
dc.contributor.authorChoo, Jaegulen_US
dc.contributor.editorAgus, Marcoen_US
dc.contributor.editorAigner, Wolfgangen_US
dc.contributor.editorHoellt, Thomasen_US
dc.date.accessioned2022-06-02T15:50:47Z
dc.date.available2022-06-02T15:50:47Z
dc.date.issued2022
dc.description.abstractImage classification models often learn to predict a class based on irrelevant co-occurrences between input features and an output class in training data. We call the unwanted correlations ''data biases,'' and the visual features causing data biases ''bias factors.'' It is challenging to identify and mitigate biases automatically without human intervention. Therefore, we conducted a design study to find a human-in-the-loop solution. First, we identified user tasks that capture the bias mitigation process for image classification models with three experts. Then, to support the tasks, we developed a visual analytics system called DASH that allows users to visually identify bias factors, to iteratively generate synthetic images using a state-of-the-art image-toimage translation model, and to supervise the model training process for improving the classification accuracy. Our quantitative evaluation and qualitative study with ten participants demonstrate the usefulness of DASH and provide lessons for future work.en_US
dc.description.sectionheadersVisual Analysis and Machine Learning
dc.description.seriesinformationEuroVis 2022 - Short Papers
dc.identifier.doi10.2312/evs.20221099
dc.identifier.isbn978-3-03868-184-7
dc.identifier.pages91-95
dc.identifier.pages5 pages
dc.identifier.urihttps://doi.org/10.2312/evs.20221099
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/evs20221099
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Human-centered computing --> Visual analytics
dc.subjectHuman centered computing
dc.subjectVisual analytics
dc.titleDASH: Visual Analytics for Debiasing Image Classification via User-Driven Synthetic Data Augmentationen_US
Files
Original bundle
Now showing 1 - 2 of 2
Loading...
Thumbnail Image
Name:
091-095.pdf
Size:
32.39 MB
Format:
Adobe Portable Document Format
No Thumbnail Available
Name:
1014-file1.mp4
Size:
34.62 MB
Format:
Unknown data format
Collections