2martens
/
uni
mirror of https://github.com/2martens/uni.git
1
0
Fork 0
Code Releases Activity

[Masterproj] Added discussion and conclusion 

Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
Jim Martens 2018-05-23 12:37:27 +02:00
parent 84974326f6
commit f230359e25
2 changed files with 69 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
masterproj/bib.bib
View File

@ -25,6 +25,16 @@
Timestamp = {2018.05.22}
}
@Article{Qi2017,
Title = {Frustum PointNets for 3D Object Detection from RGB-D Data},
Author = {Qi, Charles R and Liu, Wei and Wu, Chenxia and Su, Hao and Guibas, Leonidas J},
Journaltitle = {arXiv preprint arXiv:1711.08488},
Year = {2017},
Owner = {jim},
Timestamp = {2018.05.23}
}
@Inproceedings{Silberman2012,
Title = {Indoor segmentation and support inference from RGBD images},
Author = {Silberman, Nathan and Hoiem, Derek and Kohli, Pushmeet and Fergus, Rob},

64
masterproj/seminar_report.tex
View File

@ -318,23 +318,77 @@ information. The 2D component helps in distinguishing similar shaped objects.
\section{Discussion} % (fold)
\label{sec:discussion}
After providing the details of the paper, this secion contains your persnal opinion regarding the mothd that was proposed in this paper.
Deep Sliding Shapes offers a seemingly powerful new approach for object detection
in a 3D environment.
\subsection{Paper Strengths} % (fold)
\label{sub:paper_strengths}
Please discuss, justifying your comments with the appropriate level of details, the strengths of the paper
The paper is written in a clearly structured way and uses sub headlines to
better guide the reader. The authors apparently tried to minimize repetition
in the sentences and are using some elements of novelized storytelling like
rhetorical questions that soften up the paper and make it less dry. The introduction
in particular is giving a very good motivation for the paper and ends with a cliff
hanger that creates excitement to continue reading beyond the detour that is
the section about related works.
Overall the paper provides many illustrating figures that make it far easier
to imagine the results of the introduced method and quite simply hydrate the
paper and make it friendlier to the eyes compared to an all text paper.
Lastly the paper provides many evaluation results that are understandable
largely without the main paper text and give a good overview over the performance
of the proposed method compared to others.
% subsection positive_aspect (end)
\subsection{Paper Weaknesses} % (fold)
\label{sub:paper_weaknesses}
Please discuss, justifying your comments with the appropriate level of details, the weaknesses of the paper
That said there are things to criticize about this paper. The information about
the network structure is spread over two figures and some sections of the paper
with no guarantees that no information is missing. Furthermore no information
regarding the training, validation and testing data split were available. While
this implementation information does not have to be inside the paper proper it
should have been inside appendices to make an independent replication of results
easier. Not directly a problem with the paper itself the decision to implement
a software framework from scratch rather than using a proven existing one like
Tensorflow makes it more difficult to utilize the pretrained models which are
indeed available.
The evaluation sections are inconsistent in their structure. The first section
about object proposal evaluation follows the rest of the paper and is written
in continuous text. It describes the compared methods and then discusses the
results. The second section regarding the object detecion evaluation however
is written completely different. There is no continuous text and the compared
methods are not really described. Instead the section is largely used to justify
the chosen design. This would not even be a problem if there were a introductory
text explaining their motivations for this kind of evaluation and guiding the
reader through the process. Currently there is no explanation given why
the detection evaluation starts with feature encoding and is followed by
design justification.
Furthermore the motivations for the used data sets NYUv2 and SUN RGB-D are
not quite clear. Which data set is used for what purpose and why? The text
mentions in one sentence that the amodal bounding boxes are obtained from
SUN RGB-D without further explanation. It would have been advantageous
if the actual process of this "obtaining" were explained.
% subsection negitive (end)
% section review (end)
\section{Conclusion}
Summarize your report.
Provide some concluding discussion about the paper, along with, e.g., suggestions for future work.
Deep Sliding Shapes introduces a 3D convolutional network pipeline for
amodal 3D object detection. This pipeline consists of a regional proposal
network and a joint 2D and 3D object recognitioin network. Experimental
results show that this approach delivers better results than previous
state-of-the-art methods.
In future work this method should be compared to other 3D centric object detection
approaches like Frustum Point Net\cite{Qi2017}.
\newpage
\printbibliography