Trans2k: Unlocking the Power of Deep Models for Transparent Object Tracking
British Machine Vision Conference (BMVC) 2022
Alan Lukežič, Žiga Trojer, Jiří Matas and Matej Kristan
Visual object tracking has focused predominantly on opaque objects, while transparent object tracking received very little attention. Motivated by the uniqueness of transparent objects in that their appearance is directly affected by the background, the first dedicated evaluation dataset has emerged recently. We contribute to this effort by proposing the first transparent object tracking training dataset Trans2k that consists of over 2k sequences with 104,343 images overall, annotated by bounding boxes and segmentation masks. Noting that transparent objects can be realistically rendered by modern renderers, we quantify domain-specific attributes and render the dataset containing visual attributes and tracking situations not covered in the existing object training datasets. We observe a consistent performance boost (up to 16%) across a diverse set of modern tracking architectures when trained using Trans2k, and show insights not previously possible due to the lack of appropriate training sets.
Paper: BMVC2022
Project page: Github - including training dataset and rendering engine
BibTex citation:
@InProceedings{trans2k_bmvc2022,
Title = {Trans2k: Unlocking the Power of Deep Models for Transparent Object Tracking},
Author = {Lukezic, Alan and Trojer, Ziga and Matas, Jiri and Kristan, Matej},
Booktitle = {In Proceedings of the British Machine Vision Conference (BMVC)},
Year = {2022}
}