Fitness of General-Purpose Monocular Depth Estimation Architectures for Transparent Structures

Simple item page
dc.contributor.authorWirth, Tristanen_US
dc.contributor.authorJamili, Ariaen_US
dc.contributor.authorBuelow, Max vonen_US
dc.contributor.authorKnauthe, Volkeren_US
dc.contributor.authorGuthe, Stefanen_US
dc.contributor.editorPelechano, Nuriaen_US
dc.contributor.editorVanderhaeghe, Daviden_US
dc.date.accessioned2022-04-22T08:16:09Z
dc.date.available2022-04-22T08:16:09Z
dc.date.issued2022
dc.description.abstractDue to material properties, monocular depth estimation of transparent structures is inherently challenging. Recent advances leverage additional knowledge that is not available in all contexts, i.e., known shape or depth information from a sensor. General-purpose machine learning models, that do not utilize such additional knowledge, have not yet been explicitly evaluated regarding their performance on transparent structures. In this work, we show that these models show poor performance on the depth estimation of transparent structures. However, fine-tuning on suitable data sets, such as ClearGrasp, increases their estimation performance on the task at hand. Our evaluations show that high performance on general-purpose benchmarks translates well into performance on transparent objects after fine-tuning. Furthermore, our analysis suggests, that state-of-theart high-performing models are not able to capture a high grade of detail from both the image foreground and background at the same time. This finding shows the demand for a combination of existing models to further enhance depth estimation quality.en_US
dc.description.sectionheadersImage and Video
dc.description.seriesinformationEurographics 2022 - Short Papers
dc.identifier.doi10.2312/egs.20221020
dc.identifier.isbn978-3-03868-169-4
dc.identifier.issn1017-4656
dc.identifier.pages9-12
dc.identifier.pages4 pages
dc.identifier.urihttps://doi.org/10.2312/egs.20221020
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/egs20221020
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Computing methodologies --> Computer vision; Shape inference
dc.subjectComputing methodologies
dc.subjectComputer vision
dc.subjectShape inference
dc.titleFitness of General-Purpose Monocular Depth Estimation Architectures for Transparent Structuresen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
009-012.pdf
Size:
3.47 MB
Format:
Adobe Portable Document Format
Download
Collections
EG 2022 - Short Papers