Gradient Guided Depth Super-Resolution Using Attention-Based Cross Domain Interaction Module
Open Access
- 24 June 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 117656-117667
Abstract
Color guided depth image super-resolution (DISR), leveraging the strong structural similarity between the registered high-resolution (HR) color image and the low-resolution (LR) depth image. has achieved remarkable results. However, when the sampling factor is large, the DISR suffers from reconstructing accurate depth edges due to the severe loss of high frequency components. Crucially, direct use of the high frequency (HF) information of the registered map may transfer the high-frequency textures into the depth results and blur the depth edges. To address this, we propose GDISRNet, which is a gradient guided DISR network. It not only utilizes HF information of the color image, but also fundamentally avoids the texture copying artifacts. First, we compute gradient map from the Y channel of HR color image. Then, an Attention-based Multi-Scale Feature extraction module (AMSF) extracts the multi-scale features respectively from gradient map and the interpolated LR Depth map. Moreover, a novel Cross-Domain Interaction Module (CDIM) fuses these features to enhance depth boundaries while suppress color textures at the same time. Third, a gradient loss introduces to prevent edge smoothing effectively. Finally, a residual progressive structure of AMSF and CDIM reconstructs the SR depth image with enhanced high-frequency information. Experiments on several popular benchmark datasets including Middlebury, Sintel, Lu and NYU V2, demonstrate the superiority of the proposed GDISRNet over representative state-of-the-art methods.Keywords
Funding Information
- University Scientific Research Program of Xinjiang (XJEDU2021Y016)
- National Key Research and Development Program of China (2022ZD0115805)
- Provincial Key Science and Technology (S&T) Program of Xinjiang (2022A02011)
This publication has 53 references indexed in Scilit:
- Attention mechanisms in computer vision: A surveyComputational Visual Media, 2022
- PDR-Net: Progressive depth reconstruction network for color guided depth map super-resolutionNeurocomputing, 2022
- CODON: On Orchestrating Cross-Domain Attentions for Depth Super-ResolutionInternational Journal of Computer Vision, 2022
- Monocular 3D vehicle detection with multi-instance depth and geometry reasoning for autonomous drivingNeurocomputing, 2020
- Depth Map Super-Resolution via Multilevel Recursive Guidance and Progressive SupervisionIEEE Access, 2019
- Depth Map Super-Resolution by Deep Multi-Scale Guidance2016
- High-Resolution Stereo Datasets with Subpixel-Accurate Ground Truth2014
- A Naturalistic Open Source Movie for Optical Flow Evaluation2012
- Indoor Segmentation and Support Inference from RGBD Images2012
- A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence AlgorithmsInternational Journal of Computer Vision, 2002