Concept
Architecture d'entreprise
Publications associées (152)
Survey of Precision-Scalable Multiply-Accumulate Units for Neural-Network Processing
Christian Enz, Vincent Frédéric Camus
The current trend for deep learning has come with an enormous computational need for billions of Multiply-Accumulate (MAC) operations per inference. Fortunately, reduced precision has demonstrated large benefits with low impact on accuracy, paving the way ...
IEEE2019A Novel Approach to Modeling Enterprise Services Leveraging Object Cloning and Multilevel Classification
Object-oriented modeling is concerned with capturing common properties of objects. The dominant thinking in this realm is to classify objects that share certain properties into what is called a class, which in turn enables us to instantiate additional obje ...
IEEE2019DermoNet: densely linked convolutional neural network for efficient skin lesion segmentation
Jean-Philippe Thiran, Hazim Kemal Ekenel, Saleh Baghersalimi
Recent state-of-the-art methods for skin lesion segmentation are based on convolutional neural networks (CNNs). Even though these CNN-based segmentation approaches are accurate, they are computationally expensive. In this paper, we address this problem and ...
2019Research and development of an intelligent particle tracker detector electronic system
In high energy physics, silicon particles detectors are widely used in tracking applications. They feature high-resolution measurements of the traversing particle positioning and frequently its energy. With the new possibilities introduced by technology sc ...
EPFL2019Efficient Convolutional Neural Networks for Depth-Based Multi-Person Pose Estimation
Jean-Marc Odobez, Olivier Canévet, Michael Villamizar
Achieving robust multi-person 2D body landmark localization and pose estimation is essential for human behavior and interaction understanding as encountered for instance in HRI settings. Accurate methods have been proposed recently, but they usually rely o ...
2019Coopetition and Ecosystems: Case of Amazon.com
Alain Wegmann, Arash Golnam, Gorica Tapandjieva Sekulovska
Business and innovation ecosystems involve complex interdependencies among various actors. The concept of ecosystem is useful for analyzing strategies in which competitors are also considered as complementary partners. To make this explicit, we use SEAM – ...
Routledge2018Applications of a reconfigurable SPAD line imager
Edoardo Charbon, Claudio Bruschini, Samuel Burri
Single-photon avalanche diodes (SPADs) fabricated in standard CMOS processes enable cost-effective volume production of sensors capable of time-stamping individual photons with high accuracy. This capability makes the sensors ideally suited for application ...
2018Design and Modelling of a Super-Regenerative Receiver for Medical Implant Devices
Catherine Dehollain, Naci Pekçokgüler
Medical implant devices have been widely used in recent years. The Super-Regenerative Receiver has been on preferred architecture due to its power advantage over other architectures. We present a detailed analysis of the circuits and their equivalent model ...
IEEE2018A Row-Column Accessed Dynamic Element Matching DAC Architecture for SAR ADCs
Mustafa Kilic, Selman Ergünay, Yusuf Leblebici
Capacitor mismatch in high resolution Successive Approximation Register (SAR) Analog to Digital Converters (ADCs) causes non-linearity effects that reduce the Spurious-Free Dynamic Range (SFDR). This paper presents a novel dynamic element matching (DEM) te ...
2018Residual Parameter Transfer for Deep Domain Adaptation
Pascal Fua, Mathieu Salzmann, Artem Rozantsev
The goal of Deep Domain Adaptation is to make it possible to use Deep Nets trained in one domain where there is enough annotated training data in another where there is little or none. Most current approaches have focused on learning feature representation ...
2018