Method

DFuseNet [DFuseNet]
https://github.com/ShreyasSkandanS/DFuseNet

Submitted on 13 Nov. 2018 21:14 by
Shreyas Skandan Shivakumar (University of Pennsylvania)

Running time:0.08 s
Environment:GPU @ 2.0 Ghz (C/C++)

Method Description:
In this paper we propose a convolutional neural network that is designed to upsample a series of sparse range measurements based on the contextual cues gleaned from a high resolution intensity image. Our approach draws inspiration from related work on super-resolution and in-painting. We propose a novel architecture that seeks to pull contextual cues separately from the intensity image and the depth features and then fuse them later in the network. We argue that this approach effectively exploits the relationship between the two modalities and produces accurate results while respecting salient image structures. We present experimental results to demonstrate that our approach is comparable with state of the art methods and generalizes well across multiple datasets.
Parameters:
\learningrate=1e-4
Latex Bibtex:
@article{shivakumar2018deepfuse,
title = {DFuseNet: Deep Fusion of RGB and Sparse Depth Information for Image Guided Dense Depth Completion},
author = {Shivakumar, Shreyas S and Nguyen, Ty and Chen, Steven W. and Taylor, Camillo J},
journal = {arXiv preprint arXiv:1902.00761},
year = {2019},
}

Detailed Results

This page provides detailed results for the method(s) selected. For the first 20 test images, the percentage of erroneous pixels is depicted in the table. We use the error metric described in Sparsity Invariant CNNs (THREEDV 2017), which considers a pixel to be correctly estimated if the disparity or flow end-point error is <3px or <5% (for scene flow this criterion needs to be fulfilled for both disparity maps and the flow map). Underneath, the left input image, the estimated results and the error maps are shown (for disp_0/disp_1/flow/scene_flow, respectively). The error map uses the log-color scale described in Sparsity Invariant CNNs (THREEDV 2017), depicting correct estimates (<3px or <5% error) in blue and wrong estimates in red color tones. Dark regions in the error images denote the occluded pixels which fall outside the image boundaries. The false color maps of the results are scaled to the largest ground truth disparity values / flow magnitudes.

Test Set Average

iRMSE iMAE RMSE MAE
Error 3.62 1.79 1206.66 429.93
This table as LaTeX

Test Image 0

iRMSE iMAE RMSE MAE
Error 4.65 1.85 1646.37 476.85
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 1

iRMSE iMAE RMSE MAE
Error 5.36 2.59 1599.86 315.23
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 2

iRMSE iMAE RMSE MAE
Error 3.02 2.19 1643.66 751.67
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 3

iRMSE iMAE RMSE MAE
Error 4.27 2.62 1023.14 469.89
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 4

iRMSE iMAE RMSE MAE
Error 4.24 2.23 724.50 335.98
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 5

iRMSE iMAE RMSE MAE
Error 5.37 2.17 1299.50 401.75
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 6

iRMSE iMAE RMSE MAE
Error 7.79 2.53 881.13 356.67
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 7

iRMSE iMAE RMSE MAE
Error 6.79 2.89 1154.27 301.08
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 8

iRMSE iMAE RMSE MAE
Error 3.77 1.30 1225.49 345.14
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 9

iRMSE iMAE RMSE MAE
Error 2.97 1.67 1130.27 388.37
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 10

iRMSE iMAE RMSE MAE
Error 2.23 1.64 1016.82 540.25
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 11

iRMSE iMAE RMSE MAE
Error 3.41 1.80 1714.27 720.27
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 12

iRMSE iMAE RMSE MAE
Error 7.81 3.57 1728.21 583.87
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 13

iRMSE iMAE RMSE MAE
Error 2.03 1.28 1171.25 361.76
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 14

iRMSE iMAE RMSE MAE
Error 3.05 1.60 972.73 319.07
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 15

iRMSE iMAE RMSE MAE
Error 4.80 2.40 923.32 368.69
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 16

iRMSE iMAE RMSE MAE
Error 2.27 1.40 927.77 386.87
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 17

iRMSE iMAE RMSE MAE
Error 2.34 1.33 1096.01 375.47
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 18

iRMSE iMAE RMSE MAE
Error 3.01 1.66 982.20 469.53
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 19

iRMSE iMAE RMSE MAE
Error 1.75 1.28 1190.95 414.95
This table as LaTeX

Input Image

D1 Result

D1 Error




eXTReMe Tracker