A fantastic work in Video-level Anomaly Detection
This is my codes that can visualize the psnr image in testing videos.
Future Frame Prediction for Anomaly Detection – A New Baseline
This is a fantastic work in Video-level Anomaly Detection, published in CVPR2018. ShanghaiTech svip-lab has given their work in [Github]. Moreover, this work also have an interesting video in [YouTube]. And we can see that when anomaly examples happened, PSNR images will have a low response. Such is an example in avenue dataset.
Testing images through PSNR image on your saved models
After you have trained you pre-trained model, you need to make sure that you have done every step under the instruction of authors. You need to put videotest_psnr.py into Codes folder. Running the sript