Publication

Object Detection with Active Sample Harvesting

Related publications (54)

Sample Distillation for Object Detection and Image Classification

François Fleuret, Olivier Canévet, Leonidas Lefakis

We propose a novel approach to efficiently select informative samples for large-scale learning. Instead of directly feeding a learning algorithm with a very large amount of samples, as it is usually done to reach state-of-the-art performance, we have devel ...

2014

Sparse Binary Features for Image Classification

Johann Paratte

In this work a new method for automatic image classification is proposed. It relies on a compact representation of images using sets of sparse binary features. This work first evaluates the Fast Retina Keypoint binary descriptor and proposes imp ...

2013

Development of a low-cost wireless controller for flexible sampling strategies based on real-time flow monitoring

Andrea Rinaldo, Pierre Claude Jean Queloz, Jonathan Besuchet

Even if models are able to predict more and more accurately pollutant discharge in streams, surface water sampling remains a very common practice to monitor substance concentrations and loads in streams and to calibrate models. However, as this method is t ...

2013

Sampling and reconstructing diffusion fields in presence of aliasing

Martin Vetterli, Juri Ranieri

The reconstruction of a diffusion field, such as temperature, from samples collected by a sensor network is a classical inverse problem and it is known to be ill-conditioned. Previous work considered source models, such as sparse sources, to regularize the ...

2013

Learning to Detect Objects with Minimal Supervision

Karim Ali

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 ...

EPFL2012

Discovering single classes in remote sensing images with active learning

Devis Tuia

When dealing with supervised target detection, the acquisition of labeled samples is one of the most critical phases: the samples must be yet representative of the class of interest, but must also be found among a vast majority of non-target examples. More ...

2012

A Real-Time Deformable Detector

Pascal Fua, François Fleuret, Karim Ali, David Hasler

We propose a new learning strategy for object detection. The proposed scheme forgoes the need to train a collection of detectors dedicated to homogeneous families of poses, and instead learns a single classifier that has the inherent ability to deform base ...

Institute of Electrical and Electronics Engineers2012

Ionic liquids as matrices in microfluidic sample deposition for high-mass matrix-assisted laser desorption/ionization mass spectrometry

Henning Paul-Julius Stahlberg

Sample preparation for matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) via a microfluidic deposition device using ionic liquid matrices addresses several problems of standard protocols with crystalline matrices, such as the heterog ...

IM Publications2012

Assessment of UHPFRC based bridge rehabilitation in Slovenia, Two years after application

Emmanuel Denarié

Within the framework of the EU FP6 Project ARCHES, improved Ultra High Performance Fibre Reinforced Concretes (UHPFRC) based on local components were developed and applied to the rehabilitation of an 36 years old reinforced concrete bridge in Slovenia. Two ...

Uni Kassel2012

Tasting Families of Features for Image Classification

François Fleuret, Charles Dubout

Using multiple families of image features is a very efficient strategy to improve performance in object detection or recognition. However, such a strategy induces multiple challenges for machine learning methods, both from a computational and a statistical ...

2011

Graph Chatbot

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.