1,721,190 research outputs found
Point clouds from smartphones
No full textmartphones are omnipresent, and many people can no longer do without them. Smartphone cameras capture images suited for generating point clouds and 3D models. Apps running on smartphones and software running on a remote server enable easy 3D modelling from multiple images. The challenge is to train and guide laymen through a proper image capture strategy using their smartphones. The authors of this article investigated the potential use of smartphones for cheap and rapid generation of point clouds and 3D models exploiting a collaborative approach
Point clouds from smartphones
No full textmartphones are omnipresent, and many people can no longer do without them. Smartphone cameras capture images suited for generating point clouds and 3D models. Apps running on smartphones and software running on a remote server enable easy 3D modelling from multiple images. The challenge is to train and guide laymen through a proper image capture strategy using their smartphones. The authors of this article investigated the potential use of smartphones for cheap and rapid generation of point clouds and 3D models exploiting a collaborative approach
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Automatische analyse van visuele scenes voor autonome platformen
No full textThis thesis addresses the problem of visual scene understanding in computer vision. Automatically understanding the contents of a scene and predicting how it will evolve over time are of crucial importance for applications involving autonomous platforms like cars or drones. It is a challenging task because of the large scale and variation in training data combined with real-time constraints. We present new deep learning based methods to tackle various aspects of the scene understanding problem: semantic instance segmentation, lane detection, and video prediction.
In semantic instance segmentation, the goal is to uniquely detect, segment, and label each object in the scene. We approach this task with two methods. In the first one, a deep neural network equipped with a discriminative loss function is trained to map pixels to an embedding space such that pixel embeddings belonging to the same instance are clustered together. We explore the advantages of this embed-and-cluster approach over the popular detect-and-segment methods, and show its use in a multi-task setup. The second method relies on predicting an affinity graph of the input image, which indicates whether two pixels in a local neighborhood belong to the same object or not. In order to optimize for segmentation accuracy rather than affinity classification rate, we introduce a differentiable module that propagates a seed pixel through the predicted affinity graph, resulting in a binary segmentation mask of the object under the seed pixel. This training procedure allows us to partition the affinity graph using vanilla connected components at test time. The predicted components are treated as instance mask proposals, and are pruned and classified using region-of-interest pooling.
Apart from segmenting objects, it is often desirable to directly predict some of their parameters. In lane detection, we are ultimately interested in optimally predicting the lane line positions and curvature parameters rather than its pixel-wise segmentation mask. We present an end-to-end framework for predicting such lane parameters directly from an image, by backpropagating the loss through a weighted least-squares fitting procedure. This allows to directly optimize the network for the task of interest, without falling back on a classical two-step procedure.
With Dynamic Filter Networks, we make a contribution at the level of the network architecture. In a Dynamic Filter Network, filters are generated dynamically at inference time, conditioned on an input. They come in different variants, and are particularly well suited for tasks involving local spatial transformations. We demonstrate their effectiveness on the task of video prediction, where the goal is to predict the next frame in a video sequence. We show that the network learns to pick up flow information in a self-supervised manner, by only looking at unlabeled training data.
In order to run scene understanding algorithms on mobile platforms with a limited energy budget, neural networks must be made more energy-efficient. Our final contribution is a method for reducing the energy consumption of neural networks on embedded devices by quantizing their weights and activations after training.status: Publishe
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
