Enhancing Collaborative Road Scene Reconstruction with Unsupervised Domain Alignment

Venator M, Aklanoglu S, Bruns E, Maier A (2021)

Publication Language: English

Publication Type: Journal article

Publication year: 2021


DOI: 10.1007/s00138-020-01144-8


Scene reconstruction and visual localization in dynamic environments such as street scenes are a challenge due to the lack of distinctive, stable keypoints. While learned convolutional features have proven to be robust to changes in viewing conditions, hand-crafted features still have advantages in distinctiveness and accuracy when applied to Structure from Motion. For collaborative reconstruction of road sections by a car fleet, we apply multimodal domain adaptation as a preprocessing step to align images in their appearance and enhance keypoint matching across viewing conditions while preserving the advantages of hand-crafted features. Training a Generative Adversarial Network for translations between different illumination and weather conditions, we evaluate both qualitative and quantitative aspects of domain adaptation and its impact on feature correspondences. Combined with a multi-feature discriminator, the model is optimized for synthesis of images which do not only improve feature matching but also exhibit a high visual quality. Evaluating the approach with a challenging multi-domain image dataset recorded in various road scenes on multiple test drives, we show that our approach outperforms other traditional and learning-based methods by improving completeness or accuracy of Structure from Motion with multimodal two-domain image collections in eight out of ten test scenes.

Authors with CRIS profile

Involved external institutions

How to cite


Venator, M., Aklanoglu, S., Bruns, E., & Maier, A. (2021). Enhancing Collaborative Road Scene Reconstruction with Unsupervised Domain Alignment. Machine Vision and Applications. https://doi.org/10.1007/s00138-020-01144-8


Venator, Moritz, et al. "Enhancing Collaborative Road Scene Reconstruction with Unsupervised Domain Alignment." Machine Vision and Applications (2021).

BibTeX: Download