P
Philip Häusser
Researcher at Technische Universität München
Publications - 5
Citations - 7637
Philip Häusser is an academic researcher from Technische Universität München. The author has contributed to research in topics: Supervised learning & Optical flow. The author has an hindex of 5, co-authored 5 publications receiving 5750 citations.
Papers
More filters
Proceedings ArticleDOI
FlowNet: Learning Optical Flow with Convolutional Networks
Alexey Dosovitskiy,Philipp Fischery,Eddy Ilg,Philip Häusser,Caner Hazirbas,Vladimir Golkov,Patrick van der Smagt,Daniel Cremers,Thomas Brox +8 more
TL;DR: In this paper, the authors propose and compare two architectures: a generic architecture and another one including a layer that correlates feature vectors at different image locations, and show that networks trained on this unrealistic data still generalize very well to existing datasets such as Sintel and KITTI.
Proceedings ArticleDOI
A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation
Nikolaus Mayer,Eddy Ilg,Philip Häusser,Philipp Fischer,Daniel Cremers,Alexey Dosovitskiy,Thomas Brox +6 more
TL;DR: In this article, a large-scale synthetic stereo video dataset is proposed to enable training and evaluation of optical flow estimation with a convolutional network and disparity estimation with CNNs.
Proceedings ArticleDOI
A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation
Nikolaus Mayer,Eddy Ilg,Philip Häusser,Philipp Fischer,Daniel Cremers,Alexey Dosovitskiy,Thomas Brox +6 more
TL;DR: This paper proposes three synthetic stereo video datasets with sufficient realism, variation, and size to successfully train large networks and presents a convolutional network for real-time disparity estimation that provides state-of-the-art results.
Posted Content
FlowNet: Learning Optical Flow with Convolutional Networks
Philipp Fischer,Alexey Dosovitskiy,Eddy Ilg,Philip Häusser,Caner Hazirbas,Vladimir Golkov,Patrick van der Smagt,Daniel Cremers,Thomas Brox +8 more
TL;DR: This paper constructs CNNs which are capable of solving the optical flow estimation problem as a supervised learning task, and proposes and compares two architectures: a generic architecture and another one including a layer that correlates feature vectors at different image locations.
Posted Content
Learning by Association - A versatile semi-supervised training method for neural networks
TL;DR: In this article, a semi-supervised learning framework for deep neural networks inspired by learning in humans is proposed, where associations are made from embeddings of labeled samples to those of unlabeled ones and back.