Picture for Mathieu Salzmann

Mathieu Salzmann

Learning Monocular 3D Human Pose Estimation from Multi-view Images

Mar 24, 2018
Helge Rhodin, Jörg Spörri, Isinsu Katircioglu, Victor Constantin, Frédéric Meyer, Erich Müller, Mathieu Salzmann, Pascal Fua

Figure 1 for Learning Monocular 3D Human Pose Estimation from Multi-view Images

Figure 2 for Learning Monocular 3D Human Pose Estimation from Multi-view Images

Figure 3 for Learning Monocular 3D Human Pose Estimation from Multi-view Images

Figure 4 for Learning Monocular 3D Human Pose Estimation from Multi-view Images

* CVPR 2018, Ski-Pose PTZ-Camera Dataset available

Access Paper or Ask Questions

Compression-aware Training of Deep Networks

Nov 13, 2017
Jose M. Alvarez, Mathieu Salzmann

Figure 1 for Compression-aware Training of Deep Networks

Figure 2 for Compression-aware Training of Deep Networks

Figure 3 for Compression-aware Training of Deep Networks

Figure 4 for Compression-aware Training of Deep Networks

* Accepted at NIPS 2017

Access Paper or Ask Questions

Deep Subspace Clustering Networks

Sep 08, 2017
Pan Ji, Tong Zhang, Hongdong Li, Mathieu Salzmann, Ian Reid

Figure 1 for Deep Subspace Clustering Networks

Figure 2 for Deep Subspace Clustering Networks

Figure 3 for Deep Subspace Clustering Networks

Figure 4 for Deep Subspace Clustering Networks

* Accepted to NIPS'17

Access Paper or Ask Questions

Bringing Background into the Foreground: Making All Classes Equal in Weakly-supervised Video Semantic Segmentation

Aug 15, 2017
Fatemeh Sadat Saleh, Mohammad Sadegh Aliakbarian, Mathieu Salzmann, Lars Petersson, Jose M. Alvarez

Figure 1 for Bringing Background into the Foreground: Making All Classes Equal in Weakly-supervised Video Semantic Segmentation

Figure 2 for Bringing Background into the Foreground: Making All Classes Equal in Weakly-supervised Video Semantic Segmentation

Figure 3 for Bringing Background into the Foreground: Making All Classes Equal in Weakly-supervised Video Semantic Segmentation

Figure 4 for Bringing Background into the Foreground: Making All Classes Equal in Weakly-supervised Video Semantic Segmentation

* 11 pages, 4 figures, 7 tables, Accepted in ICCV 2017

Access Paper or Ask Questions