Unit
Environmental Computational Science and Earth Observation Laboratory
LaboratoryRelated publications (63)
Large-Scale Image Segmentation with Convolutional Networks
Pedro Henrique Oliveira Pinheiro
Object recognition is one of the most important problems in computer vision. However, visual recognition poses many challenges when tried to be reproduced by artificial systems. A main challenge is the problem of variability: objects can appear across huge ...
Sciences et Techniques de l’Ingénieur (STI)2017Word Sequence Modeling using Deep Learning
For a long time, natural language processing (NLP) has relied on generative models with task specific and manually engineered features. Recently, there has been a resurgence of interest for neural networks in the machine learning community, obtaining state ...
EPFL2016Theory of representation learning in cortical neural networks
Our brain continuously self-organizes to construct and maintain an internal representation of the world based on the information arriving through sensory stimuli. Remarkably, cortical areas related to different sensory modalities appear to share the same f ...
EPFL2016Combining Human Computing and Machine Learning to Make Sense of Big (Aerial) Data for Disaster Response
Stéphane Joost, Devis Tuia, Matthew Josef Parkan, Carlos Castillo, Muhammad Imran, Patrick Meier, Julien Briant
Aerial imagery captured via unmanned aerial vehicles (UAVs) is playing an increasingly important role in disaster response. Unlike satellite imagery, aerial imagery can be captured and processed within hours rather than days. In addition, the spatial resol ...
Mary Ann Liebert, Inc2016Innovations for protecting ridge and reef: a CSR strategy and practice
This paper will describe and analyze the innovations adopted by a Filipino businessman and his family corporation in order to protect the forest and rehabilitate coral and marine life. The first innovation involved the engagement of villagers in forest pro ...
Villigen PSI, World Resources Forum, printed by Paul Scherrer Institute2015Semi-supervised and unsupervised kernel-based novelty detection with application to remote sensing images
Frank Grégoire Jean de Morsier
The main challenge of new information technologies is to retrieve intelligible information from the large volume of digital data gathered every day. Among the variety of existing data sources, the satellites continuously observing the surface of the Earth ...
EPFL2014Learning to Detect Objects with Minimal Supervision
Many classes of objects can now be successfully detected with statistical machine learning techniques. Faces, cars and pedestrians, have all been detected with low error rates by learning their appearance in a highly generic manner from extensive training ...
EPFL2012Natural Language Processing (Almost) from Scratch
Ronan Collobert, Michael Karlen
We propose a unified neural network architecture and learning algorithm that can be applied to various natural language processing tasks including part-of-speech tagging, chunking, named entity recognition, and semantic role labeling. This versatility is a ...
2011Identifying Conservation and Restoration Priorities for Saproxylic and Old-Growth Forest Species: A Case Study in Switzerland
Saproxylic (dead-wood-associated) and old-growth species are among the most threatened species in European forest ecosystems, as they are susceptible to intensive forest management. Identifying areas with particular relevant features of biodiversity is of ...
Springer-Verlag2009Semi-automatic Use of High Resolution Images and Digital Elevation Models for Counting and Identification of Forest Trees
This paper presents a semi-automatic and systematic computational approach intending to count and localize different species of trees in zones of dense forest. Comparative analysis of the application of multi-spectral high resolution images and aerial LIDA ...
2007