1,720,969 research outputs found
Decoding human dynamics: explorations in motion forecasting, social navigation, and egocentric perception
Full text linkHuman dynamics—how individuals move, interact, and perceive their environment—pose significant challenges for theoretical understanding and practical implementation in robotics, human-computer interaction, and behavior analysis. Accurate models addressing these challenges are essential for developing intelligent systems capable of effectively collaborating with or understanding humans. This Ph.D. thesis investigates
key aspects of human dynamics through Motion Forecasting, Social Navigation, and Egocentric Perception. In Motion Forecasting, we explore both two-body pose prediction and global human motion prediction. We present best practices for improving collaborative motion prediction [261]. We introduce a staged, contact-aware framework for global human motion forecasting [282] that predicts human movements within
broader environmental contexts. Our model surpasses existing methods by incorporating contact points and staged motion, enabling more accurate human pose and trajectory predictions. In the context of social dynamics, we investigate the impact of latent variables on forecasting human interactions, especially in team-based settings. Introducing a role-based approach demonstrates that understanding these latent social roles can significantly improve trajectory prediction in multi-agent systems [281]. This concept extends to Social Navigation [280], where a robot’s trajectory planning must
account for human movement and be processed in real-time. Human dynamics are incorporated into the robot’s reinforcement learning path-planning framework via a social dynamics module. This module distills human trajectories into latent codes, which serve as contextual input for the robot’s policy model. We also address challenges in Egocentric Perception and Mistake Detection. By developing a novel method, we tackle the need for real-time online detection of procedural mistakes from egocentric video streams. Our approach, PREGO [93], introduces an innovative model that recognizes current actions and predicts future ones to identify discrepancies and detect mistakes. We also present an extension of the latter, which offers an in-depth analysis and enhances the framework with an Automatic Chain of Thought mechanism. This addition improves the model’s reasoning capabilities, enabling more nuanced error detection. Additionally, we contribute a framework for estimating social interactions and human meshes using egocentric video, improving pose estimation accuracy by incorporating wearer-interactee interactions.
Beyond direct applications to human dynamics, this thesis includes a contribution to Topological Deep Learning. We contributed to a technical paper introducing the first Python framework for Topological Deep Learning [119], offering new tools for researchers exploring machine learning on non-Euclidean data structures. Overall, this thesis explores human motion forecasting, social interaction modeling, and egocentric perception while advancing methodologies in machine learning. The insights and tools developed contribute to understanding human behavior and pave the way for further research in intelligent systems and interactive environments
Mesoscale precipitation nowcasting from weather radar data using space-time-separable graph convolutional networks
No full textBest Practices for 2-Body Pose Forecasting
No full textThe task of collaborative human pose forecasting stands for predicting the future poses of multiple interacting people, given those in previous frames. Predicting two people in interaction, instead of each separately, promises better performance, due to their body-body motion correlations. But the task has remained so far primarily unexplored.In this paper, we review the progress in human pose forecasting and provide an in-depth assessment of the single-person practices that perform best for 2-body collaborative motion forecasting. Our study confirms the positive impact of frequency input representations, space-time separable and fully-learnable interaction adjacencies for the encoding GCN and FC decoding. Other single-person practices do not transfer to 2-body, so the proposed best ones do not include hierarchical body modeling or attention-based interaction encoding.We further contribute a novel initialization procedure for the 2-body spatial interaction parameters of the encoder, which benefits performance and stability. Altogether, our proposed 2-body pose forecasting best practices yield a performance improvement of 21.9% over the state-of-the-art on the most recent ExPI dataset, whereby the novel initialization accounts for 3.5%. See our project page at https://www.pinlab.org/bestpractices2bod
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
About latent roles in forecasting players in team sports
Full text linkForecasting players in sports has grown in popularity due to the potential
for a tactical advantage and the applicability of such research to multi-agent
interaction systems. Team sports contain a significant social component that
influences interactions between teammates and opponents. However, it still
needs to be fully exploited. In this work, we hypothesize that each participant
has a specific function in each action and that role-based interaction is
critical for predicting players' future moves. We create RolFor, a novel
end-to-end model for Role-based Forecasting. RolFor uses a new module we
developed called Ordering Neural Networks (OrderNN) to permute the order of the
players such that each player is assigned to a latent role. The latent role is
then modeled with a RoleGCN. Thanks to its graph representation, it provides a
fully learnable adjacency matrix that captures the relationships between roles
and is subsequently used to forecast the players' future trajectories.
Extensive experiments on a challenging NBA basketball dataset back up the
importance of roles and justify our goal of modeling them using optimizable
models. When an oracle provides roles, the proposed RolFor compares favorably
to the current state-of-the-art (it ranks first in terms of ADE and second in
terms of FDE errors). However, training the end-to-end RolFor incurs the issues
of differentiability of permutation methods, which we experimentally review.
Finally, this work restates differentiable ranking as a difficult open problem
and its great potential in conjunction with graph-based interaction models.
Project is available at: https://www.pinlab.org/aboutlatentrole
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
- …
