Deep-DSO: Improving Mapping of Direct Sparse Odometry Using CNN-Based Single-Image Depth Estimation

Erick P. Herrera-Granda (), Juan C. Torres-Cantero, Israel D. Herrera-Granda, José F. Lucio-Naranjo, Andrés Rosales, Javier Revelo-Fuelagán and Diego H. Peluffo-Ordóñez ()
Additional contact information
Erick P. Herrera-Granda: Department of Informatics and Computer Science, Escuela Politécnica Nacional, Quito 170525, Ecuador
Juan C. Torres-Cantero: Virtual Reality Laboratory, ETSIIT, Department of Computer Languages and Systems, University of Granada, 18071 Granada, Spain
Israel D. Herrera-Granda: Administration and Business Economics, Foreign Trade Program, Faculty of International Trade and Integration, Universidad Politécnica Estatal del Carchi, Calle Antisana y Av. Universitaria, Tulcán 040102, Ecuador
José F. Lucio-Naranjo: Department of Informatics and Computer Science, Escuela Politécnica Nacional, Quito 170525, Ecuador
Andrés Rosales: Departamento de Automatización y Control Industrial, GIECAR, Escuela Politécnica Nacional, Quito 170525, Ecuador
Javier Revelo-Fuelagán: Department of Electronic Engineering, Universidad de Nariño, Pasto 52001, Colombia
Diego H. Peluffo-Ordóñez: SDAS Research Group, Ben Guerir 43150, Morocco

Mathematics, 2025, vol. 13, issue 20, 1-28

Abstract: In recent years, SLAM, visual odometry, and structure-from-motion approaches have widely addressed the problems of 3D reconstruction and ego-motion estimation. Of the many input modalities that can be used to solve these ill-posed problems, the pure visual alternative using a single monocular RGB camera has attracted the attention of multiple researchers due to its low cost and widespread availability in handheld devices. One of the best proposals currently available is the Direct Sparse Odometry (DSO) system, which has demonstrated the ability to accurately recover trajectories and depth maps using monocular sequences as the only source of information. Given the impressive advances in single-image depth estimation using neural networks, this work proposes an extension of the DSO system, named DeepDSO. DeepDSO effectively integrates the state-of-the-art NeW CRF neural network as a depth estimation module, providing depth prior information for each candidate point. This reduces the point search interval over the epipolar line. This integration improves the DSO algorithm’s depth point initialization and allows each proposed point to converge faster to its true depth. Experimentation carried out in the TUM-Mono dataset demonstrated that adding the neural network depth estimation module to the DSO pipeline significantly reduced rotation, translation, scale, start-segment alignment, end-segment alignment, and RMSE errors.

Keywords: CNN direct sparse odometry; monocular visual odometry; monocular 3D reconstruction; monocular ego-motion; pure visual odometry (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/13/20/3330/pdf (application/pdf)
https://www.mdpi.com/2227-7390/13/20/3330/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:13:y:2025:i:20:p:3330-:d:1774837

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().