Author
Jean-Philippe Thiran
Other affiliations: University of Lausanne
Bio: Jean-Philippe Thiran is an academic researcher from École Polytechnique Fédérale de Lausanne. The author has contributed to research in topics: Convolutional neural network & Normalization (image processing). The author has an hindex of 10, co-authored 43 publications receiving 1115 citations. Previous affiliations of Jean-Philippe Thiran include University of Lausanne.
Papers
More filters
••
German Cancer Research Center1, Université de Sherbrooke2, University Health Network3, University of Pittsburgh4, IMT Institute for Advanced Studies Lucca5, St. Jude Children's Research Hospital6, University of Toronto7, Zhejiang University of Technology8, Harvard University9, Utrecht University10, Université de Montréal11, National Research Council12, University of Washington13, University of Western Ontario14, École Polytechnique Fédérale de Lausanne15, ETSI16, Siemens17, University of Southern California18, King's College London19, University of Bordeaux20, Centre national de la recherche scientifique21, Copenhagen University Hospital22, University of Hamburg23, University of Basel24
TL;DR: The encouraging finding that most state-of-the-art algorithms produce tractograms containing 90% of the ground truth bundles (to at least some extent) is reported, however, the same tractograms contain many more invalid than valid bundles, and half of these invalid bundles occur systematically across research groups.
Abstract: Tractography based on non-invasive diffusion imaging is central to the study of human brain connectivity. To date, the approach has not been systematically validated in ground truth studies. Based on a simulated human brain data set with ground truth tracts, we organized an open international tractography challenge, which resulted in 96 distinct submissions from 20 research groups. Here, we report the encouraging finding that most state-of-the-art algorithms produce tractograms containing 90% of the ground truth bundles (to at least some extent). However, the same tractograms contain many more invalid than valid bundles, and half of these invalid bundles occur systematically across research groups. Taken together, our results demonstrate and confirm fundamental ambiguities inherent in tract reconstruction based on orientation information alone, which need to be considered when interpreting tractography and connectivity results. Our approach provides a novel framework for estimating reliability of tractography and encourages innovation to address its current limitations.
996 citations
••
01 Jan 2019TL;DR: This work presents a new dataset for form understanding in noisy scanned documents (FUNSD) that aims at extracting and structuring the textual content of forms, and is the first publicly available dataset with comprehensive annotations to address FoUn task.
Abstract: We present a new dataset for form understanding in noisy scanned documents (FUNSD) that aims at extracting and structuring the textual content of forms. The dataset comprises 199 real, fully annotated, scanned forms. The documents are noisy and vary widely in appearance, making form understanding (FoUn) a challenging task. The proposed dataset can be used for various tasks, including text detection, optical character recognition, spatial layout analysis, and entity labeling/linking. To the best of our knowledge, this is the first publicly available dataset with comprehensive annotations to address FoUn task. We also present a set of baselines and introduce metrics to evaluate performance on the FUNSD dataset, which can be downloaded at https://guillaumejaume.github.io/FUNSD.
190 citations
••
01 Oct 2019TL;DR: In this paper, the authors optimize a deep network-based decoder with a targeted objective function that penalizes images at different semantic levels using the corresponding terms, which results in more realistic textures and sharper edges.
Abstract: By benefiting from perceptual losses, recent studies have improved significantly the performance of the super-resolution task, where a high-resolution image is resolved from its low-resolution counterpart. Although such objective functions generate near-photorealistic results, their capability is limited, since they estimate the reconstruction error for an entire image in the same way, without considering any semantic information. In this paper, we propose a novel method to benefit from perceptual loss in a more objective way. We optimize a deep network-based decoder with a targeted objective function that penalizes images at different semantic levels using the corresponding terms. In particular, the proposed method leverages our proposed OBB (Object, Background and Boundary) labels, generated from segmentation labels, to estimate a suitable perceptual loss for boundaries, while considering texture similarity for backgrounds. We show that our proposed approach results in more realistic textures and sharper edges, and outperforms other state-of-the-art algorithms in terms of both qualitative results on standard benchmarks and results of extensive user studies.
113 citations
••
German Cancer Research Center1, Université de Sherbrooke2, University Health Network3, Carnegie Mellon University4, IMT Institute for Advanced Studies Lucca5, St. Jude Children's Research Hospital6, Zhejiang University of Technology7, Harvard University8, Utrecht University9, Université de Montréal10, National Research Council11, University of Washington12, University of Western Ontario13, École Polytechnique Fédérale de Lausanne14, ETSI15, Siemens16, University of Southern California17, King's College London18, University of Bordeaux19, French Institute for Research in Computer Science and Automation20, Copenhagen University Hospital21, University of Hamburg22, University Hospital of Basel23
TL;DR: The results demonstrate fundamental ambiguities inherent to tract reconstruction methods based on diffusion orientation information, with critical consequences for the approach of diffusion tractography in particular and human connectivity studies in general.
Abstract: Fiber tractography based on non-invasive diffusion imaging is at the heart of connectivity studies of the human brain. To date, the approach has not been systematically validated in ground truth studies. Based on a simulated human brain dataset with ground truth white matter tracts, we organized an open international tractography challenge, which resulted in 96 distinct submissions from 20 research groups. While most state-of-the-art algorithms reconstructed 90% of ground truth bundles to at least some extent, on average they produced four times more invalid than valid bundles. About half of the invalid bundles occurred systematically in the majority of submissions. Our results demonstrate fundamental ambiguities inherent to tract reconstruction methods based on diffusion orientation information, with critical consequences for the approach of diffusion tractography in particular and human connectivity studies in general.
76 citations
••
01 Aug 2017TL;DR: In this article, LiDAR data is utilized to generate region proposals by processing the three dimensional point cloud that it provides and these candidate regions are then further processed by a state-of-the-art CNN classifier that has been fine-tuned for pedestrian detection.
Abstract: Pedestrian detection is an important component for safety of autonomous vehicles, as well as for traffic and street surveillance. There are extensive benchmarks on this topic and it has been shown to be a challenging problem when applied on real use-case scenarios. In purely image-based pedestrian detection approaches, the state-of-the-art results have been achieved with convolutional neural networks (CNN) and surprisingly few detection frameworks have been built upon multi-cue approaches. In this work, we develop a new pedestrian detector for autonomous vehicles that exploits LiDAR data, in addition to visual information. In the proposed approach, LiDAR data is utilized to generate region proposals by processing the three dimensional point cloud that it provides. These candidate regions are then further processed by a state-of-the-art CNN classifier that we have fine-tuned for pedestrian detection. We have extensively evaluated the proposed detection process on the KITTI dataset. The experimental results show that the proposed LiDAR space clustering approach provides a very efficient way of generating region proposals leading to higher recall rates and fewer misses for pedestrian detection. This indicates that LiDAR data can provide auxiliary information for CNN-based approaches.
41 citations
Cited by
More filters
•
1,261 citations
01 Jan 2016
TL;DR: This book helps people to enjoy a good book with a cup of coffee in the afternoon, instead they juggled with some malicious bugs inside their laptop.
Abstract: Thank you for downloading magnetic resonance imaging physical principles and sequence design. As you may know, people have look numerous times for their chosen books like this magnetic resonance imaging physical principles and sequence design, but end up in harmful downloads. Rather than enjoying a good book with a cup of coffee in the afternoon, instead they juggled with some malicious bugs inside their laptop.
695 citations
••
TL;DR: In this article, the authors systematically summarize methodologies and discuss challenges for deep multi-modal object detection and semantic segmentation in autonomous driving and provide an overview of on-board sensors on test vehicles, open datasets, and background information for object detection.
Abstract: Recent advancements in perception for autonomous driving are driven by deep learning. In order to achieve robust and accurate scene understanding, autonomous vehicles are usually equipped with different sensors (e.g. cameras, LiDARs, Radars), and multiple sensing modalities can be fused to exploit their complementary properties. In this context, many methods have been proposed for deep multi-modal perception problems. However, there is no general guideline for network architecture design, and questions of “what to fuse”, “when to fuse”, and “how to fuse” remain open. This review paper attempts to systematically summarize methodologies and discuss challenges for deep multi-modal object detection and semantic segmentation in autonomous driving. To this end, we first provide an overview of on-board sensors on test vehicles, open datasets, and background information for object detection and semantic segmentation in autonomous driving research. We then summarize the fusion methodologies and discuss challenges and open questions. In the appendix, we provide tables that summarize topics and methods. We also provide an interactive online platform to navigate each reference: https://boschresearch.github.io/multimodalperception/ .
674 citations
••
Charité1, Wayne State University2, Ottawa Hospital Research Institute3, University of Luxembourg4, University of Melbourne5, University of Freiburg6, Radboud University Nijmegen7, University of South Carolina8, University of Pittsburgh9, Northeastern University10, Harvard University11, University of Utah12
TL;DR: This work represents a multi‐institutional collaborative effort to develop a comprehensive, open source pipeline for DBS imaging and connectomics, which has already empowered several studies, and may facilitate a variety of future studies in the field.
473 citations