mirror of
https://github.com/tu-darmstadt-informatik/bsc-thesis.git
synced 2026-02-06 10:05:40 +00:00
WIP
This commit is contained in:
parent
a6311dca56
commit
582de90668
48
bib.bib
48
bib.bib
@ -31,11 +31,11 @@
|
||||
journal = {arXiv preprint arXiv:1704.07804},
|
||||
year = {2017}}
|
||||
|
||||
@article{MaskRCNN,
|
||||
@inproceedings{MaskRCNN,
|
||||
Author = {Kaiming He and Georgia Gkioxari and
|
||||
Piotr Doll\'{a}r and Ross Girshick},
|
||||
Title = {{Mask {R-CNN}}},
|
||||
Journal = {arXiv preprint arXiv:1703.06870},
|
||||
Booktitle = {CVPR},
|
||||
Year = {2017}}
|
||||
|
||||
@inproceedings{FasterRCNN,
|
||||
@ -80,19 +80,19 @@
|
||||
@article{VGGNet,
|
||||
author = {Karen Simonyan and Andrew Zisserman},
|
||||
title = {Very Deep Convolutional Networks for Large-Scale Image Recognition},
|
||||
journal = {arXiv preprint arXiv:1409.1556},
|
||||
year = {2014}}
|
||||
journal = {ICLR},
|
||||
year = {2015}}
|
||||
|
||||
@article{ResNet,
|
||||
@inproceedings{ResNet,
|
||||
author = {Kaiming He and Xiangyu Zhang and Shaoqing Ren and Jian Sun},
|
||||
title = {Deep Residual Learning for Image Recognition},
|
||||
journal = {arXiv preprint arXiv:1512.03385},
|
||||
year = {2015}}
|
||||
booktitle = {CVPR},
|
||||
year = {2016}}
|
||||
|
||||
@article{DenseNetDenseFlow,
|
||||
author = {Yi Zhu and Shawn D. Newsam},
|
||||
title = {DenseNet for Dense Flow},
|
||||
journal = {arXiv preprint arXiv:1707.06316},
|
||||
journal = {ICIP},
|
||||
year = {2017}}
|
||||
|
||||
@inproceedings{SE3Nets,
|
||||
@ -114,10 +114,10 @@
|
||||
year = {2016}}
|
||||
|
||||
@inproceedings{VKITTI,
|
||||
author = {Adrien Gaidon and Qiao Wang and Yohann Cabon and Eleonora Vig},
|
||||
title = {Virtual Worlds as Proxy for Multi-Object Tracking Analysis},
|
||||
booktitle = {{CVPR}},
|
||||
year = {2016}}
|
||||
author = {Adrien Gaidon and Qiao Wang and Yohann Cabon and Eleonora Vig},
|
||||
title = {Virtual Worlds as Proxy for Multi-Object Tracking Analysis},
|
||||
booktitle = {{CVPR}},
|
||||
year = {2016}}
|
||||
|
||||
@inproceedings{KITTI2012,
|
||||
author = {Andreas Geiger and Philip Lenz and Raquel Urtasun},
|
||||
@ -152,14 +152,14 @@
|
||||
@article{SPyNet,
|
||||
author = {Anurag Ranjan and Michael J. Black},
|
||||
title = {Optical Flow Estimation using a Spatial Pyramid Network},
|
||||
journal = {arXiv preprint arXiv:1611.00850},
|
||||
year = {2016}}
|
||||
journal = {CVPR},
|
||||
year = {2017}}
|
||||
|
||||
@article{FPN,
|
||||
@inproceedings{FPN,
|
||||
author = {Tsung-Yi Lin and Piotr Dollár and Ross Girshick and Kaiming He and Bharath Hariharan and Serge Belongie},
|
||||
title = {Feature Pyramid Networks for Object Detection},
|
||||
journal = {arXiv preprint arXiv:1612.03144},
|
||||
year = {2016}}
|
||||
booktitle = {CVPR},
|
||||
year = {2017}}
|
||||
|
||||
@inproceedings{PoseNet,
|
||||
author = {Alex Kendall and Matthew Grimes and Roberto Cipolla},
|
||||
@ -190,3 +190,17 @@
|
||||
title = {Deeper Depth Prediction with Fully Convolutional Residual Networks},
|
||||
booktitle = {3DV},
|
||||
year = {2016}}
|
||||
|
||||
@inproceedings{TensorFlowObjectDetection,
|
||||
author = {J. Huang and V. Rathod and C. Sun and M. Zhu and A. Korattikara and A. Fathi and I. Fischer and Z. Wojna,
|
||||
and Y. Song and S. Guadarrama and K. Murphy},
|
||||
title = {Speed/accuracy trade-offs for modern convolutional object detectors},
|
||||
booktitle = {CVPR},
|
||||
year = {2017}}
|
||||
|
||||
@misc{TensorFlow,
|
||||
title={{TensorFlow}: Large-Scale Machine Learning on Heterogeneous Systems},
|
||||
url={http://tensorflow.org/},
|
||||
note={Software available from tensorflow.org},
|
||||
author={Martín Abadi and others},
|
||||
year={2015}}
|
||||
|
||||
@ -1,3 +1,18 @@
|
||||
\subsection{Implementation}
|
||||
Our networks and loss functions are implemented using built-in TensorFlow \cite{TensorFlow}
|
||||
functions, enabling us to use automatic differentiation for all gradient
|
||||
computations. To make our code easy to extend and flexible, we build on
|
||||
the TensorFlow Object detection API \cite{TensorFlowObjectDetection}, which provides a Faster R-CNN baseline
|
||||
implementation.
|
||||
On top of this, we implemented Mask R-CNN and the Feature Pyramid Network (FPN)
|
||||
as well all extensions for motion estimation and related evaluations
|
||||
and postprocessings. In addition, we generated all ground truth for
|
||||
Motion R-CNN in the form of TFRecords from the raw Virtual KITTI
|
||||
data to enable fast loading during training.
|
||||
Note that for RoI pooling and cropping,
|
||||
we use the \texttt{tf.crop\_and\_resize} TensorFlow function with
|
||||
interpolation set to bilinear.
|
||||
|
||||
\subsection{Datasets}
|
||||
|
||||
\paragraph{Virtual KITTI}
|
||||
|
||||
@ -54,7 +54,7 @@
|
||||
url=false,
|
||||
hyperref=auto,
|
||||
giveninits=true,
|
||||
maxbibnames=10
|
||||
maxbibnames=100
|
||||
]{biblatex}
|
||||
|
||||
\renewbibmacro*{cite:seenote}{} % um zu verhindern, dass in Fußnoten automatisch "(wie Anm. xy)" eingefügt wird
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user