diff options
Diffstat (limited to 'report2')
| -rwxr-xr-x | report2/metadata.yaml | 3 | ||||
| -rwxr-xr-x | report2/paper.md | 7 |
2 files changed, 5 insertions, 5 deletions
diff --git a/report2/metadata.yaml b/report2/metadata.yaml index 20375f9..5500c63 100755 --- a/report2/metadata.yaml +++ b/report2/metadata.yaml @@ -6,6 +6,9 @@ author: numbersections: yes lang: en babel-lang: english +nocite: | + @deepreid, @sklearn + abstract: | This report analyses distance metrics learning techniques with regards to identification accuracy for the dataset CUHK03. The baseline method used for diff --git a/report2/paper.md b/report2/paper.md index 6358445..20bd053 100755 --- a/report2/paper.md +++ b/report2/paper.md @@ -130,7 +130,7 @@ repository complimenting this paper. The approach addressed to improve the identification performance is based on k-reciprocal reranking. The following section summarizes the idea behind -the method illustrated in **REFERENCE PAPER**. +the method illustrated in reference @rerank-paper. We define $N(p,k)$ as the top k elements of the ranklist generated through NN, where p is a query image. The k reciprocal ranklist, $R(p,k)$ is defined as the @@ -185,7 +185,7 @@ in steps of 0.1. The results obtained through this approach suggest: $k_{1_{opt} Reranking achieves better results than the other baseline methods analyzed both as $top k$ accuracy and mean average precision. It is also necessary to estimate how precise the ranklist generated is. -For this reason an additional method of evaluation is introduced: mAP. +For this reason an additional method of evaluation is introduced: mAP. See reference @mAP. It is possible to see in figure \ref{fig:ranklist2} how the ranklist generated for the same five queries of figure \ref{fig:eucrank} has improved for the fifth query. The mAP improves from 47% to 61.7%. @@ -237,6 +237,3 @@ training are close to the ones for the local maximum of gallery and query. # References -# Appendix - - |