Skip to content

Commit

Permalink
typo and remove TODO
Browse files Browse the repository at this point in the history
  • Loading branch information
Sproc01 committed Jul 21, 2024
1 parent 37f99b0 commit e17cf83
Show file tree
Hide file tree
Showing 4 changed files with 2 additions and 5 deletions.
Binary file modified Output/game2_clip1_output.mp4
Binary file not shown.
2 changes: 1 addition & 1 deletion Report/section/results.tex
Original file line number Diff line number Diff line change
Expand Up @@ -71,7 +71,7 @@ \subsection{Quantitative results}
This leads to a lower overall mean Average Precision (mAP) that is focused only on balls, but a good mean Intersection over Union (mIoU) because the background and playing field are well segmented.

\subsection{Qualitative results}
ome qualitative results are presented below.
Some qualitative results are presented below.

% TOOD descrivere un minimo qualitativamente le immagini con qualcosa di vero su tutto il dataset (osservazioni sul funzionamento/qualità)

Expand Down
2 changes: 1 addition & 1 deletion Report/section/tableSegmentation.tex
Original file line number Diff line number Diff line change
Expand Up @@ -6,5 +6,5 @@ \subsection{Table segmentation}
\item[Color-based mask] Created by identifying pixels corresponding to the table's color range. This mask is very robust on shadows and reflections since it defined through the Hue component of the HSV color space, which is invariant to brightness.
\item[k-means clustering mask] Generated by applying the k-means algorithm on the image to separate the table from the background. Useful when the table color range is too broad to be captured by the color-based mask.
\end{description}
These two masks re combined through a voting system, which classifies as table pixels classified as such by at least one of the two masks gaining the best of both approaches.
These two masks are combined through a voting system, which classifies as table pixels classified as such by at least one of the two masks gaining the best of both approaches.
Finally the isolated area is limited to the pixels inside the polygon defined by the previously detected corners.
3 changes: 0 additions & 3 deletions src/main.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -44,9 +44,6 @@ int main(int argc, char *argv[]) {
if (argc == 2) {
videoPath = filesystem::path(argv[1]);
}
else if (argc == 1) { //TODO: remove at the end
videoPath = filesystem::path("../Dataset/game2_clip1/game2_clip1.mp4");
}
else {
cout << "Error of number of parameters: insert one parameter" << endl;
return -1;
Expand Down

0 comments on commit e17cf83

Please sign in to comment.