@proceedings {6299, title = {Divide and Conquer the Embedding Space for Metric Learning}, year = {2019}, keywords = {deep learning, metric learning}, url = {https://github.com/CompVis/metric-learning-divide-and-conquer}, author = {Sanakoyeu, A. and Tschernezki, V. and Uta B{\"u}chler and Bj{\"o}rn Ommer} } @proceedings {6300, title = {Using a Transformation Content Block For Image Style Transfer}, year = {2019}, author = {Dmytro Kotovenko and Sanakoyeu, A. and Sabine Lang and Ma, P. and Bj{\"o}rn Ommer} } @article {6229, title = {Deep Unsupervised Learning of Visual Similarities}, journal = {Pattern Recognition}, volume = {78}, year = {2018}, chapter = {331}, abstract = {Exemplar learning of visual similarities in an unsupervised manner is a problem of paramount importance to Computer Vision. In this context, however, the recent breakthrough in deep learning could not yet unfold its full potential. With only a single positive sample, a great imbalance between one positive and many negatives, and unreliable relationships between most samples, training of Convolutional Neural networks is impaired. In this paper we use weak estimates of local similarities and propose a single optimization problem to extract batches of samples with mutually consistent relations. Conflicting relations are distributed over different batches and similar samples are grouped into compact groups. Learning visual similarities is then framed as a sequence of categorization tasks. The CNN then consolidates transitivity relations within and between groups and learns a single representation for all samples without the need for labels. The proposed unsupervised approach has shown competitive performance on detailed posture analysis and object classification.}, doi = {https://doi.org/10.1016/j.patcog.2018.01.036}, url = {https://authors.elsevier.com/a/1WXUt77nKSb25 }, author = {Sanakoyeu, A. and Miguel Bautista and Bj{\"o}rn Ommer} } @conference {style_aware_content_loss_eccv18, title = {A Style-Aware Content Loss for Real-time HD Style Transfer}, booktitle = {Proceedings of the European Conference on Computer Vision (ECCV) (Oral)}, year = {2018}, abstract = {Recently, style transfer has received a lot of attention. While much of this research has aimed at speeding up processing, the approaches are still lacking from a principled, art historical standpoint: a style is more than just a single image or an artist, but previous work is limited to only a single instance of a style or shows no benefit from more images. Moreover, previous work has relied on a direct comparison of art in the domain of RGB images or on CNNs pre-trained on ImageNet, which requires millions of labeled object bounding boxes and can introduce an extra bias, since it has been assembled without artistic consideration. To circumvent these issues, we propose a style-aware content loss, which is trained jointly with a deep encoder-decoder network for real-time, high-resolution stylization of images and videos. We propose a quantitative measure for evaluating the quality of a stylized image and also have art historians rank patches from our approach against those from previous work. These and our qualitative results ranging from small image patches to megapixel stylistic images and videos show that our approach better captures the subtle nature in which a style affects content. }, keywords = {deep learning, generative network, Style transfer}, author = {Sanakoyeu, A. and Dmytro Kotovenko and Sabine Lang and Bj{\"o}rn Ommer} } @conference {6200, title = {Deep Unsupervised Similarity Learning using Partially Ordered Sets}, booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)}, year = {2017}, author = {Miguel Bautista and Sanakoyeu, A. and Bj{\"o}rn Ommer} } @conference {arXiv:1608.08792, title = {CliqueCNN: Deep Unsupervised Exemplar Learning}, booktitle = {Proceedings of the Conference on Advances in Neural Information Processing Systems (NIPS)}, year = {2016}, publisher = {MIT Press}, organization = {MIT Press}, address = {Barcelona}, abstract = {Exemplar learning is a powerful paradigm for discovering visual similarities in an unsupervised manner. In this context, however, the recent breakthrough in deep learning could not yet unfold its full potential. With only a single positive sample, a great imbalance between one positive and many negatives, and unreliable relationships between most samples, training of Convolutional Neural networks is impaired. Given weak estimates of local distance we propose a single optimization problem to extract batches of samples with mutually consistent relations. Conflict- ing relations are distributed over different batches and similar samples are grouped into compact cliques. Learning exemplar similarities is framed as a sequence of clique categorization tasks. The CNN then consolidates transitivity relations within and between cliques and learns a single representation for all samples without the need for labels. The proposed unsupervised approach has shown competitive performance on detailed posture analysis and object classification.}, url = {https://arxiv.org/abs/1608.08792}, author = {Miguel Bautista and Sanakoyeu, A. and Sutter, E. and Bj{\"o}rn Ommer} }