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

Fixed dashes

Browse Source 
Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
Jim Martens
2019-09-24 15:38:41 +02:00
parent 43820194c8
commit cb92f63775
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
body.tex
View File

@ -443,7 +443,7 @@ SSD network are the predictions with class confidences, offsets to the
anchor box, anchor box coordinates, and variance. The model loss is a anchor box, anchor box coordinates, and variance. The model loss is a
weighted sum of localisation and confidence loss. As the network weighted sum of localisation and confidence loss. As the network
has a fixed number of anchor boxes, every forward pass creates the same has a fixed number of anchor boxes, every forward pass creates the same
number of detections - 8732 in the case of SSD 300x300. number of detections---8732 in the case of SSD 300x300.
Notably, the object proposals are made in a single run for an image - Notably, the object proposals are made in a single run for an image -
single shot. single shot.
@ -960,8 +960,8 @@ averaging was not reported in their paper.
There is no visible impact of entropy thresholding on the object detection There is no visible impact of entropy thresholding on the object detection
performance for vanilla SSD. This indicates that the network has almost no performance for vanilla SSD. This indicates that the network has almost no
uniform or close to uniform predictions, the vast majority of predictions uniform or close to uniform predictions, the vast majority of predictions
has a high confidence in one class - including the background. has a high confidence in one class---including the background.
However, the entropy plays a larger role for the Bayesian variants - as However, the entropy plays a larger role for the Bayesian variants---as
expected: the best performing thresholds are 1.0, 1.3, and 1.4 for micro averaging, expected: the best performing thresholds are 1.0, 1.3, and 1.4 for micro averaging,
and 1.5, 1.7, and 2.0 for macro averaging. In all of these cases the best and 1.5, 1.7, and 2.0 for macro averaging. In all of these cases the best
threshold is not the largest threshold tested. A lower threshold likely threshold is not the largest threshold tested. A lower threshold likely
@ -986,7 +986,7 @@ have the same number of observations everywhere before the entropy threshold. Af
Without NMS 79\% of observations are left. Irrespective of the absolute Without NMS 79\% of observations are left. Irrespective of the absolute
number, this discrepancy clearly shows the impact of non-maximum suppression and also explains a higher count of false positives: number, this discrepancy clearly shows the impact of non-maximum suppression and also explains a higher count of false positives:
more than 50\% of the original observations were removed with NMS and more than 50\% of the original observations were removed with NMS and
stayed without - all of these are very likely to be false positives. stayed without---all of these are very likely to be false positives.
A clear distinction between micro and macro averaging can be observed: A clear distinction between micro and macro averaging can be observed:
recall is hardly effected with micro averaging (0.300) but goes down equally with macro averaging (0.229). For micro averaging, it does recall is hardly effected with micro averaging (0.300) but goes down equally with macro averaging (0.229). For micro averaging, it does