Bienvenidos a LIV
Página en proceso.
Nuestro Equipo
Ver todos los miembros →
María Elena Buemi
Investigadora UBA
Daniel Acevedo
Investigador CONICET
Pablo Negri
Investigador CONICET
Últimas Novedades
Ver todas las novedades →Presentación de poster en el Third Joint Egocentric Vision (EgoVis) Workshop (CVPR2026) por Juan Ignacio Bustos Gorostegui
Ver la novedad completaLo que las redes neuronales no saben: descubriendo e interpretando el conocimiento aprendido en redes neuronales convolucionales.
Este trabajo aborda el problema de inferir el conocimiento adquirido por una red neuronal convolucional para clasificación de imágenes, sin información previa sobre sus datos de entrenamiento. En particular, se estudia si es posible determinar qué clases fueron utilizadas...
Artículos Destacados
Explorar todos los artículos →Event-based Liveness Detection using Temporal Ocular Dynamics: An Exploratory Approach
Face liveness detection has been extensively studied using RGB cameras, achieving strong performance under controlled conditions but often failing to generalize across sensors and attack scenarios. In this work, we explore event cameras as an alternative sensing modality for liveness detection based on temporal ocular dynamics...
Leer artículo completoClimate Change and Integrated Resource Management: A Multi-Scale Modeling Framework for Water and Food Security in Argentina
This paper presents the construction and an application of SIMPLE-G-AR, the first spatially explicit, gridded partial equilibrium model of agricultural production and water use calibrated for Argentina. Argentina is a major global food exporter whose agricultural sector is highly exposed to climate variability: only 5.2 percent of cropland relies on artificial irrigation, while recurrent droughts such as the 2022–2023 La Niña event caused export declines of 38 percent in primary products and 27 percent in agro-industrial manufactures...
Leer artículo completoCross-Modal Action Recognition in Egocentric Video Using Mamba: Integrating RGB and Hand Skeleton Streams via CLS Token Fusion Strategies
Egocentric action recognition is a challenging task due to erratic camera motion, frequent hand occlusion, and the difficulty of maintaining consistent visual representations over time. In this work, we propose a cross-modal architecture that combines RGB video and temporal hand skeleton data within a unified Mamba-based framework, exploiting the linear time complexity of State Space Models (SSMs)...
Leer artículo completo