Original scientific paper
https://doi.org/10.30765/er.3383
Deep recurrent visual place recognition based on softmax fine-tuned image representations
Jurica Maltar
orcid.org/0000-0002-9078-6347
; J. J. Strossmayer University of Osijek, School of Applied Mathematics and Informatics, Osijek, Croatia
*
Ivan Marković
; University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, Croatia
Domagoj Matijević
orcid.org/0000-0003-3390-9467
; J. J. Strossmayer University of Osijek, School of Applied Mathematics and Informatics, Osijek, Croatia
Ivan Petrović
orcid.org/0000-0001-9961-5627
; University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, Croatia
* Corresponding author.
Abstract
Visual place recognition aims to identify previously visited locations from image representations that are robust to viewpoint and environmental changes. Deep convolutional neural network features perform well for this task, yet they are typically pretrained for classification or segmentation – objectives that do not account for the sequential nature of place recognition. In mobile robotics, exploiting image sequences can further improve performance. We propose softmax-based fine-tuning of a convolutional network extended with a recurrent model to enhance place recognition on image sequences. Experiments on two public datasets show that the proposed representation consistently outperforms competing approaches across all evaluated methods.
Keywords
visual place recognition; deep convolutional neural networks; recurrent neural networks; softmax regression; SeqSLAM
Hrčak ID:
349926
URI
Publication date:
29.7.2026.
Visits: 0 *