2026-06-01T10:16:03Zhttps://riuma.uma.es/rest/oai/request

oai:riuma.uma.es:10630/327382026-02-03T10:58:49Zcom_10630_2254col_10630_37953

Marín-Jiménez, Manuel J. Castro Payán, Francisco Manuel Carmona-Poyato, Ángel Guil-Mata, Nicolás 2024-09-20T10:44:00Z 2024-09-20T10:44:00Z 2015-12-15 Manuel J. Marín-Jiménez, Francisco M. Castro, Ángel Carmona-Poyato, Nicolás Guil, On how to improve tracklet-based gait recognition systems, Pattern Recognition Letters, Volume 68, Part 1, 2015, Pages 103-110, ISSN 0167-8655, https://doi.org/10.1016/j.patrec.2015.08.025. https://hdl.handle.net/10630/32738 https://doi.org/10.1016/j.patrec.2015.08.025 Abstract Recently, short-term dense trajectories features (DTF) have shown state-of-the-art results in video recognition and retrieval. However, their use has not been extensively studied on the problem of gait recognition. Therefore, the goal of this work is to propose and evaluate diverse strategies to improve recognition performance in the task of gait recognition based on DTF. In particular, this paper will show that (i) the proposed RootDCS descriptor improves on DCS in most tested cases; (ii) selecting relevant trajectories in an automatic way improves the recognition performance in several situations; (iii) applying a metric learning technique to reduce dimensionality of feature vectors improves on standard PCA; and, (iv) binarization of low-dimensionality feature vectors not only reduces storage needs but also improves recognition performance in many cases. The experiments are carried out on the popular datasets CASIA, parts B and C, and TUM-GAID showing improvement on state-of-the-art results for most scenarios. eng http://creativecommons.org/licenses/by-nc-nd/4.0/ open access Attribution-NonCommercial-NoDerivatives 4.0 Internacional Reconocimiento de formas (Informática) Hombre - Identificación On how to improve tracklet-based gait recognition systems journal article