Skip to main content

ImmersiveDepth: A Hybrid Approach for Monocular Depth Estimation from 360 Images Using Tangent Projection and Multi-Model Integration.

Dorosti, S. and Yang, X., 2025. ImmersiveDepth: A Hybrid Approach for Monocular Depth Estimation from 360 Images Using Tangent Projection and Multi-Model Integration. In: O’Conner, L., ed. 2025 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW). New York, NY: IEEE, 1392-1393.

Full text available as:

[thumbnail of ImmersiveDepth(Final).pdf]
Preview
PDF
ImmersiveDepth(Final).pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.

691kB

DOI: 10.1109/VRW66409.2025.00342

Abstract

ImmersiveDepth is a hybrid framework designed to tackle challenges in Monocular Depth Estimation (MDE) from 360-degree images, specifically spherical distortions, occlusions, and texture inconsistencies. By integrating tangent image projection, a combination of convolutional neural networks (CNNs) and transformer models, and a novel multi-scale alignment process, ImmersiveDepth achieves seamless and precise depth predictions. Evaluations on diverse datasets show an average 37% reduction in RMSE compared to Depth Anything V2 and a 25% accuracy boost in low-light conditions over MiDaS v3.1. ImmersiveDepth thus establishes a robust solution for immersive technologies, autonomous systems, and 3D reconstruction.

Item Type:Book Section
Additional Information:08-12 March 2025, Saint Malo, France
Uncontrolled Keywords:Monocular depth estimation; 360-degree images; tangent projection; VR; AR; SfM; MVS
Group:Faculty of Media & Communication
ID Code:41147
Deposited By: Symplectic RT2
Deposited On:02 Jul 2025 14:57
Last Modified:02 Jul 2025 14:57

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -