From 616484fb8bf8803a0e74f4c68843b63f2a384703 Mon Sep 17 00:00:00 2001 From: Vasil Zlatanov Date: Mon, 10 Dec 2018 17:10:21 +0000 Subject: Improvements to current sections --- report2/paper.md | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) diff --git a/report2/paper.md b/report2/paper.md index e4c3ec0..dc07241 100755 --- a/report2/paper.md +++ b/report2/paper.md @@ -2,22 +2,16 @@ ## Probelm Definition -The problem to solve is to create a ranklist for each image of the query set -by finding the nearest neighbor(s) within a gallery set. However gallery images -with the same label and taken from the same camera as the query image should -not be considered when forming the ranklist. +The person re-identification problem presented in this paper requires mtatching pedestrian images from disjoint camera's by pedestrian detectors. This problem is challenging, as identities captured in photsos are subject to various lighting, pose, blur, background and oclusion from various camera views. This report considers features extracted from the CUHK03 dataset, following a 50 layer Residual network (Resnet50),. This paper considers distance metrics techniques which can be used to perform person re-identification across **disjoint* cameras, using these features. ## Dataset - CUHK03 The dataset CUHK03 contains 14096 pictures of people captured from two -different cameras. The feature vectors used come from passing the -rescaled images through ResNet50. Each feature vector contains 2048 -features that we use for classification. The pictures represent 1467 different -people and each of them appears between 9 and 10 times. The separation of -train_idx, query_idx and gallery_idx allows to perform taining and validation -on a training set (train_idx, adequately split between test, train and -validation keeping the same number of identities). This prevents overfitting -the algorithm to the specific data associated with query_idx and gallery_idx. +different cameras. The feature vectors used, extracted from a trained ResNet50 model +, contain 2048 features that are used for identification. + +The pictures represent 1467 different +identities, each of which appears 9 to 10 times. Data is seperated in train, query and gallery sets with `train_idx`, `query_idx` and `gallery_idx` respectively, where the training set has been used to develop the ResNet50 model used for feature extraction. This procedure has allowed the evaluation of distance metric learning techniques on the query and gallery sets, without an overfit feature set a the set, as it was explicitly trained on the training set. ## Nearest Neighbor ranklist @@ -25,10 +19,13 @@ Nearest Neighbor aims to find the gallery image whose feature are the closest to the ones of a query image, predicting the class of the query image as the same of its nearest neighbor(s). The distance between images can be calculated through different distance metrics, however one of the most commonly used is euclidean -distance, represented as $d=\sqrt{\sum (x-y)^{2}}$. +distance: + +$$ NN(x) \operatorname*{argmin}_{i\in[m]} \|x-x_i\|^2 $$ -EXPLAIN KNN BRIEFLY +*Square root when calculating euclidean distance is ommited as it does not affect ranking by distance* +Alternative distance metrics exist such as jaccardian and mahalanobis, which can be used as an alternative to euclidiean distance. # Baseline Evaluation -- cgit v1.2.3-54-g00ecf From bbb2f0825f62842784c332ef316ffca947b368be Mon Sep 17 00:00:00 2001 From: Vasil Zlatanov Date: Mon, 10 Dec 2018 17:13:06 +0000 Subject: Add baseline intro --- report2/paper.md | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/report2/paper.md b/report2/paper.md index dc07241..2173cf1 100755 --- a/report2/paper.md +++ b/report2/paper.md @@ -29,6 +29,8 @@ Alternative distance metrics exist such as jaccardian and mahalanobis, which can # Baseline Evaluation +To evaluate improvements brought by alternative distance learning metrics a baseline is established as trough nearest neighbour identification as previously described. + \begin{figure} \begin{center} \includegraphics[width=20em]{fig/baseline.pdf} @@ -55,12 +57,6 @@ Alternative distance metrics exist such as jaccardian and mahalanobis, which can \end{center} \end{figure} - -TODO: -~~ -s/kNN/NN/ -~~ - \begin{figure} \begin{center} \includegraphics[width=20em]{fig/comparison.pdf} -- cgit v1.2.3-54-g00ecf From 70b329dfaf54a09f913d29b8e602999eb56be169 Mon Sep 17 00:00:00 2001 From: Vasil Zlatanov Date: Mon, 10 Dec 2018 17:18:30 +0000 Subject: Add comments on Mahal and add sections --- report2/paper.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/report2/paper.md b/report2/paper.md index 2173cf1..0f5c9ae 100755 --- a/report2/paper.md +++ b/report2/paper.md @@ -49,6 +49,11 @@ To evaluate improvements brought by alternative distance learning metrics a base # Suggested Improvement +## kMean Clustering + + +## k-reciprocal Reranking + \begin{figure} \begin{center} \includegraphics[width=24em]{fig/ranklist.png} @@ -73,6 +78,9 @@ To evaluate improvements brought by alternative distance learning metrics a base \end{center} \end{figure} +# Comment on Mahalnobis Distance as a metric + +We were not able to achieve significant improvements using mahalanobis for original distance ranking compared to square euclidiaen metrics. Results can be observed using the `-m|--mahalanobis` when running evalution with the repository complimenting this paper. # Conclusion -- cgit v1.2.3-54-g00ecf From c031478576e1e1dbc285e2d7815c6f772ab00a01 Mon Sep 17 00:00:00 2001 From: Vasil Zlatanov Date: Mon, 10 Dec 2018 17:19:57 +0000 Subject: Improve NN equation --- report2/paper.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/report2/paper.md b/report2/paper.md index 0f5c9ae..bd559d7 100755 --- a/report2/paper.md +++ b/report2/paper.md @@ -21,7 +21,7 @@ of its nearest neighbor(s). The distance between images can be calculated throug different distance metrics, however one of the most commonly used is euclidean distance: -$$ NN(x) \operatorname*{argmin}_{i\in[m]} \|x-x_i\|^2 $$ +$$ \textrm{NN}(x) = \operatorname*{argmin}_{i\in[m]} \|x-x_i\|^2 $$ *Square root when calculating euclidean distance is ommited as it does not affect ranking by distance* -- cgit v1.2.3-54-g00ecf From 2a5c62f9ea50971ba25c3e8f519e224093ec0090 Mon Sep 17 00:00:00 2001 From: Vasil Zlatanov Date: Mon, 10 Dec 2018 17:21:38 +0000 Subject: s/propose/evaluate/ --- report2/metadata.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/report2/metadata.yaml b/report2/metadata.yaml index f35d6aa..20375f9 100755 --- a/report2/metadata.yaml +++ b/report2/metadata.yaml @@ -10,7 +10,7 @@ abstract: | This report analyses distance metrics learning techniques with regards to identification accuracy for the dataset CUHK03. The baseline method used for identification is Eucdidian based Nearest Neighbors based on Euclidean distance. - The improved approach we propose utilises Jaccardian metrics to rearrange the NN + The improved approach evaluated utilises Jaccardian metrics to rearrange the NN ranklist based on reciprocal neighbours. While this approach is more complex and introduced new hyperparameter, significant accuracy improvements are observed - approximately 10% increased Top-1 identifications, and good improvements for Top-$N$ accuracy with low $N$. ... -- cgit v1.2.3-54-g00ecf