2martens/masterthesis-latex
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Updated abstract with more info

Browse Source 
Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
Jim Martens
2019-09-25 12:52:25 +02:00
parent 9e26e4ce64
commit 5ef52d5fa9
1 changed files with 8 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
abstract.tex
View File

@ -3,12 +3,14 @@
Object detection in open set conditions is an important part of Object detection in open set conditions is an important part of
applying object detection to real world data sets. The key question applying object detection to real world data sets. The key question
is how to identify unknown or novel data. This thesis compares is how to identify unknown or novel data. Dropout sampling with entropy
vanilla SSD and SSD with dropout sampling on the MS COCO data set. thresholding is one possible avenue to answer the question.
The hyper parameters confidence threshold, entropy threshold,
usage of non-maximum suppression, and usage of dropout layers during This thesis compares vanilla SSD and SSD with dropout sampling on the MS COCO
testing are varied to create different scenarios. Both results from macro and data set. The hyper parameters confidence threshold, entropy threshold,
micro averaging are presented. usage of non-maximum suppression, usage of dropout layers during
testing, and the dropout keep ratio are varied to create different scenarios.
Both results from macro and micro averaging are presented.
Results show a significant improvement of the object detection performance Results show a significant improvement of the object detection performance
with a higher confidence threshold. The usage of non-maximum suppression with a higher confidence threshold. The usage of non-maximum suppression