Langanzeige der Metadaten
DC Element | Wert | Sprache |
---|---|---|
dc.contributor.advisor | Krueger, Joerg | - |
dc.contributor.author | Lehr, Jan | - |
dc.contributor.author | Chavan, Vivek | - |
dc.contributor.author | Koch, Paul | - |
dc.contributor.author | Schlüter, Marian | - |
dc.contributor.author | Briese, Clemens | - |
dc.date.accessioned | 2023-08-21T09:58:07Z | - |
dc.date.available | 2023-06-19T11:59:32Z | - |
dc.date.available | 2023-08-21T09:58:07Z | - |
dc.date.issued | 2023 | - |
dc.identifier.uri | https://fordatis.fraunhofer.de/handle/fordatis/329.2 | - |
dc.identifier.uri | http://dx.doi.org/10.24406/fordatis/266.2 | - |
dc.description.abstract | The Industrial Objects in Varied Contexts (InVar) dataset was internally produced by our team and contains 100 objects in 20800 total images (208 images per class). The objects consist of common automotive, machine and robotics lab parts. Each class contains 4 sub-categories (52 images each) with different attributes and visual complexities. White background (Dwh): The object is against a clean white background and the object is clear, centred and in focus. Stationary Setup (Dst): These images are also taken against a clean background using a stationary camera setup, with uncentered objects at a constant distance. The images have lower DPI resolution with occasional cropping. Handheld (Dha): These images are taken with the user holding the objects, with occasional occluding. Cluttered background (Dcl): These images are taken with the object placed along with other objects from the lab in the background and with no occlusion. The dataset was produced to simulate the miscellaneous issues in industrial setups as discussed. The dataset was produced by our staff at different workstations and labs in Berlin. More details regarding the objects used for digitisation are available in the metadata file. | en |
dc.language.iso | en | en |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | en |
dc.subject | industrial objects | en |
dc.subject | objects in context | en |
dc.subject | fine-grained classification | en |
dc.subject | multimodal dataset | en |
dc.subject | occlusion | en |
dc.subject | clutter | en |
dc.subject | computer vision | en |
dc.subject | pattern recognition | en |
dc.subject.ddc | DDC::600 Technik, Medizin, angewandte Wissenschaften::620 Ingenieurwissenschaften | en |
dc.title | InVar-100: Industrial Objects in Varied Contexts Dataset | en |
dc.type | Image | en |
dc.relation.issupplementto | Chavan, V., Koch, P., Schlüter, M., Briese, C. (2023). Towards Realistic Evaluation of Industrial Continual Learning Scenarios with an Emphasis on Energy Consumption and Computational Footprint, to appear in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) | - |
fordatis.group | Produktion | en |
fordatis.institute | IPK Institut für Produktionsanlagen und Konstruktionstechnik | en |
fordatis.rawdata | false | en |
Enthalten in den Sammlungen: | Institut für Produktionsanlagen und Konstruktionstechnik IPK |
Dateien zu dieser Ressource:
Datei | Beschreibung | Größe | Format | |
---|---|---|---|---|
InVar_100_Class_Metadata.csv | 11,17 kB | CSV | Öffnen/Download | |
image_data.zip | 30,76 GB | ZIP | Öffnen/Download | |
InVar-100_Datasheet_V2.pdf | 6,64 MB | Adobe PDF | Öffnen/Download |
Versionshistorie
Version | Ressource | Datum | Zusammenfassung |
---|---|---|---|
2 | fordatis/329.2 | 2023-08-14 11:46:31.857 | The data repository has been updated for the camera-ready submission of the ICCV 2023 paper: Towards Realistic Evaluation of Industrial Continual Learning Scenarios with an Emphasis on Energy Consumption and Computational Footprint. |
1 | fordatis/329 | 2023-06-19 13:59:32.0 | Please use only the updated data under https://fordatis.fraunhofer.de/handle/fordatis/329.2. |
Diese Ressource wurde unter folgender Copyright-Bestimmung veröffentlicht: Lizenz von Creative Commons