Method for person re-identification based on deep model with multi-loss fusion training strategy
The invention relates to a method for person re-identification based on deep model with multi-loss fusion training strategy. The method uses a deep learning technology to perform preprocessing operations such as flipping, clipping, random erasing and style transfer, and then feature extraction is performed through a backbone network model; joint training of a network is performed by fusing a plurality of loss functions. Compared with other deep learning-based person re-identification algorithms, the present invention greatly improves the performance of person re-identification by adopting a plurality of preprocessing modes, the fusion of three loss functions and effective training strategy.
1. A method for person re-identification based on a deep model with multi-loss fusion training strategy, comprising the following steps:
1): Acquiring an original image data set, dividing the original image data set into a training set and a test set, and dividing the test set into a query set and a gallery set;
2): Sequentially subjecting image data of the training set to a data preprocessing process of flipping, noise adding, automatic clipping, random erasing and style transfer, and performing data augmentation after the preprocessing is completed;
3): Selecting and training a benchmark network, updating the weight, optimizing the benchmark network, and adjusting a hyper-parameter; wherein, the benchmark network is trained by a fusion of triplet loss function, cross-entropy loss function and center loss function, wherein the triplet loss function is used to increase an inter-class distance and shorten an intra-class distance, and the center loss function is used to make feature maps of the same identity (ID) close to the center;
4): Inputting the training set image data obtained in step 2) into the optimized and adjusted benchmark network for feature extraction; and
5): Calculating Euclidean distances for the extracted features in pairs, sorting the calculated Euclidean distances, and selecting from the gallery set an image closest to a target in the query set as an identification result.
2. The method for person re-identification based on a deep model with multi-loss fusion training strategy according to claim 1 , wherein in step 3), for a large number of person re-identification data set, the transfer learning method is adopted to initialize pre-trained model parameters, and then training is further performed; and for a person data set with a small amount of data, the model trained on the large data is used to fine-tune the training.
3. The method for person re-identification based on a deep model with multi-loss fusion training strategy according to claim 1 , wherein a triplet model is used as a skeleton and three images are used as a group of inputs, and an expression of a group of input images is:
R i =<R i o ,R i + ,R i o >
wherein R i o , R i + and R i − are expressions of the group of three images in the input, respectively, R i o and R i + are positive sample pairs, respectively, and R i o and R i − are negative sample pairs, respectively.
4. The method for person re-identification based on a deep model with multi-loss fusion training strategy according to claim 1 , wherein an expression of the loss function L for fusion of the triplet loss function, the cross entropy loss function and the center loss function is:
L=α 1 L 1 +α 2 L 2 +α 3 L 3
wherein L 1 is the cross-entropy loss function, L 2 is the triplet loss function, L 3 is the center loss function, α 1 is the weight of a proportion of the cross-entropy loss, α 2 is the weight of a proportion of the triplet loss, and α 3 is the weight of a proportion of the center loss.
5. The method for person re-identification based on a deep model with multi-loss fusion training according to claim 4 , wherein the expression of the cross-entropy loss function L 1 is:
L
1
=
-
1
m
∑
i
=
1
m
log
e
W
y
i
T
x
i
+
b
yi
∑
j
=
1
n
e
W
j
T
x
i
+
b
j
wherein i is the sample, x i is the feature class, m represents the sample batch size, n represents the number of person IDs, b j is the j-th offset value, y i is label information of an input i-th sample, W i is the i-th column of the parameter matrix W of the feature layer, and T represents a transpose of parameter matrix.
6. The method for person re-identification based on deep model with multi-loss fusion training strategy according to claim 5 , wherein the corresponding expression of the triplet loss function L 2 is:
L 2 =[thre+ d )( F w ( R i o ), F w ( R i + ))− d ( F w ( R i o ), F w ( R i − ))] +
wherein thre is a hyper-parameter used to make a distance between sample pairs of the same class smaller than a distance between sample pairs of different classes, d(.) represents the distance measurement function, F w (R i o ), F w (R i + ) and F w (R i − ) are feature maps corresponding to R i o , R i + and R i − respectively, and [x] + is a function max (0, x).
7. The method for person re-identification based on deep model with multi-loss fusion training strategy according to claim 6 , wherein the corresponding expression of the center loss function L 3 is:
L
3
=
1
2
m
∑
i
=
1
m
x
i
-
center
y
i
2
2
wherein center yi is a center point of the feature with x i class being y i .
8. The method for person re-identification based on deep model with multi-loss fusion training strategy according to claim 7 , wherein the center of the center loss function L 3 is continuously updated in the training process, and the update formula is as follows when s=y i :
Δ
cente
r
s
t
=
∑
i
=
1
m
center
s
-
x
i
m
+
1
center
s
t
+
1
=
center
j
t
-
λ
·
Δ
center
s
t
wherein λ is a learning rate of the center loss function and Δ is an increment of a variable.
9. The method for person re-identification based on deep model with multi-loss fusion training strategy according to claim 1 , wherein in Step 3), adjusting the hyper-parameters of the benchmark network comprises iteration step adjustment, initial value adjustment of iteration step, and selection of learning functions.