Seite - 86 - in Joint Austrian Computer Vision and Robotics Workshop 2020

[26] H. Touvron, A. Vedaldi, M. Douze, and H. Je´gou. Fixing the train-test resolution discrepancy. ArXiv, 2019. 1 [27] S. Xie, R. B. Girshick, P. Dolla´r, Z. Tu, and K. He. Aggregated residual transformations fordeepneural networks. In IEEE Conference on Computer Vision andPatternRecognition(CVPR), pages5987–5995, 2016. 2 [28] F.YuandV.Koltun. Multi-scalecontextaggregation bydilatedconvolutions. ArXiv, 2016. 2 [29] X. Zhang, X. Zhou, M. Lin, and J. Sun. Shufflenet: Anextremelyefficientconvolutionalneuralnetwork for mobile devices. In IEEE Conference on Com- puterVisionandPatternRecognition(CVPR),pages 6848–6856, 2018. 2 86