Awesome
Online Anomaly Detection in Surveillance Videos with Asymptotic Bounds on False Alarm Rate
Implementation in Jupyter, Python 3.6, tensorflow.
0. All the files along with the checkpoints can be downloaded from http://doi.org/10.5281/zenodo.3676032. Any changes to the order of files in the directory might raise errors so please follow the uploaded file structure. To get the final results, you can directly run MONAD.ipynb. To train models from scratch, following are the instructions to run the code.
1. Installation (Anaconda with python3.6 installation is recommended)
- Install 3rd-package dependencies of python (listed in requirements.txt)
numpy==1.14.1
scipy==1.0.0
matplotlib==2.1.2
tensorflow-gpu==1.4.1
tensorflow==1.4.1
Pillow==5.0.0
pypng==0.0.18
scikit_learn==0.19.1
opencv-python==3.2.0.6
2. Installation of Darknet (YOLOv3)
Please follow the instructions provided at https://pjreddie.com/darknet/yolo/
3. Get Nominal Mean Square Error
- Running the sript (as ped2 for examples) and cd into Codes folder at first.
python inference.py --dataset ped2 \
--test_folder ../Data/ped2/training/frames \
--gpu 1 \
--snapshot_dir checkpoints/pretrains/ped2
The executed results are saved in ped2_train.
## 4. Get Testing Mean Square Error
* Running the sript (as ped2 for examples) and cd into **Codes** folder at first.
```shell
python inference.py --dataset ped2 \
--test_folder ../Data/ped2/testing/frames \
--gpu 1 \
--snapshot_dir checkpoints/pretrains/ped2
The executed results are saved in ped2_test.
## 5. Training from scratch (here we use ped2 and avenue datasets for examples)
* Set hyper-parameters
The default hyper-parameters, such as $\lambda_{init}$, $\lambda_{gd}$, $\lambda_{op}$, $\lambda_{adv}$ and the learning rate of G, as well as D, are all initialized in **training_hyper_params/hyper_params.ini**.
* Running script (as ped2 or avenue for instances) and cd into **Codes** folder at first.
```shell
python train.py --dataset ped2 \
--train_folder ../Data/ped2/training/frames \
--test_folder ../Data/ped2/testing/frames \
--gpu 0 \
--iters 80000
- Model selection while training In order to do model selection, a popular way is to testing the saved models after a number of iterations or epochs (Since there are no validation set provided on above all datasets, and in order to compare the performance with other methods, we just choose the best model on testing set). Here, we can use another GPU to listen the snapshot_dir folder. When a new model.cpkt.xxx has arrived, then load the model and test. Finnaly, we choose the best model. Following is the script.
python inference.py --dataset ped2 \
--test_folder ../Data/ped2/testing/frames \
--gpu 1
Run python train.py -h to know more about the flag options or see the detials in constant.py.
Options to run the network.
optional arguments:
-h, --help show this help message and exit
-g GPU, --gpu GPU the device id of gpu.
-i ITERS, --iters ITERS
set the number of iterations, default is 1
-b BATCH, --batch BATCH
set the batch size, default is 4.
--num_his NUM_HIS set the time steps, default is 4.
-d DATASET, --dataset DATASET
the name of dataset.
--train_folder TRAIN_FOLDER
set the training folder path.
--test_folder TEST_FOLDER
set the testing folder path.
--config CONFIG the path of training_hyper_params, default is
training_hyper_params/hyper_params.ini
--snapshot_dir SNAPSHOT_DIR
if it is folder, then it is the directory to save
models, if it is a specific model.ckpt-xxx, then the
system will load it for testing.
--summary_dir SUMMARY_DIR
the directory to save summaries.
--psnr_dir PSNR_DIR the directory to save psnrs results in testing.
--evaluate EVALUATE the evaluation metric, default is compute_auc
6. Run Object Detection
Run YOLOv3 on the training and testing datasets. Save the bounding boxes and extracted classes along with their confidence levels. Saved bounding boxes are also uploaded as json files.
7. Run MONAD
The proposed sequential detector is implemented in MONAD.ipynb. Use Anaconda 3.6 to acheive the final AUC score.