BridgeNet: A Joint Learning Network of Depth Map Super-Resolution and Monocular Depth Estimation

  title={BridgeNet: A Joint Learning Network of Depth Map Super-Resolution and Monocular Depth Estimation},
  author={Qi Tang and Runmin Cong and Ronghui Sheng and Lingzhi He and Dan Zhang and Yao-Dong Zhao and Sam Tak Wu Kwong},
  journal={Proceedings of the 29th ACM International Conference on Multimedia},
  • Qi Tang, Runmin Cong, +4 authors S. Kwong
  • Published 2021
  • Computer Science
  • Proceedings of the 29th ACM International Conference on Multimedia
Depth map super-resolution is a task with high practical application requirements in the industry. Existing color-guided depth map super-resolution methods usually necessitate an extra branch to extract high-frequency detail information from RGB image to guide the low-resolution depth map reconstruction. However, because there are still some differences between the two modalities, direct information transmission in the feature dimension or edge map dimension cannot achieve satisfactory result… Expand

Figures and Tables from this paper


Hierarchical Features Driven Residual Learning for Depth Map Super-Resolution
A novel deep network for depth map super-resolution (SR), called DepthSR-Net, built on residual U-Net deep network architecture that automatically infers a high-resolution depth map from its low-resolution version by hierarchical features driven residual learning. Expand
Residual dense network for intensity-guided depth map enhancement
A novel DCNN is proposed to progressively reconstruct the high-resolution depth map guided by the intensity image, where the multi-scale intensity features are extracted to provide guidance for the refinement of depth features as their resolutions are gradually enhanced. Expand
Learning Scene Structure Guidance via Cross-Task Knowledge Transfer for Single Depth Super-Resolution
An auxiliary depth estimation task is constructed that takes an RGB image as input to estimate a depth map, and both DSR task and DE task collaboratively to boost the performance of DSR, and a cross-task interaction module is proposed to realize bilateral cross- task knowledge transfer. Expand
Depth Map Super-Resolution by Deep Multi-Scale Guidance
A new method to address the problem of depth map super resolution in which a high-resolution (HR) depth map is inferred from a LR depth map and an additional HR intensity image of the same scene is presented. Expand
PMBANet: Progressive Multi-Branch Aggregation Network for Scene Depth Super-Resolution
A progressive multi-branch aggregation network (PMBANet), which consists of stacked MBA blocks to fully address the above problems and progressively recover the degraded depth map. Expand
Multi-Direction Dictionary Learning Based Depth Map Super-Resolution With Autoregressive Modeling
A compound method is proposed that combines multi-direction dictionary sparse representation and autoregressive (AR) models, so that the depth edges are presented precisely at different levels, and outperforms state-of-the-art depth map super-resolution methods in terms of both quantitative metrics and subjective visual quality. Expand
Deeper Depth Prediction with Fully Convolutional Residual Networks
A fully convolutional architecture, encompassing residual learning, to model the ambiguous mapping between monocular images and depth maps is proposed and a novel way to efficiently learn feature map up-sampling within the network is presented. Expand
CLIFFNet for Monocular Depth Estimation with Hierarchical Embedding Loss
A hierarchical loss for monocular depth estimation, which measures the differences between the prediction and ground truth in hierarchical embedding spaces of depth maps, and a cross level identity feature fusion network (CLIFFNet), where low-level features with finer details are refined using more reliable high-level cues. Expand
Towards Fast and Accurate Real-World Depth Super-Resolution: Benchmark Dataset and Baseline
This work constructs a large-scale dataset named “RGB-D-D”, in which the high-frequency component adaptively decomposed from RGB image to guide the depth map SR, and provides a fast depth map super-resolution (FDSR) baseline. Expand
Depth Super-Resolution with Deep Edge-Inference Network and Edge-Guided Depth Filling
Experimental results show that this method outperforms the state-of-art methods in both the edges inference and the final results of depth super-resolution, and generalizes well for handling depth data captured in different scenes. Expand