ktr-hubrt / wsal Goto Github PK
View Code? Open in Web Editor NEWOfficial codes for paper: Localizing Anomalies from Weakly-Labeled Videos
Official codes for paper: Localizing Anomalies from Weakly-Labeled Videos
Hello! what a great work it is!
I'm using your datset, TAD. But I have a question.
In your test dataset(val dataset), Normal_218.mp4 is not normal video.
There is a car accident starting from 243.jpg.
I know normal video data in this area has no abnormal scene, I wonder why you seperate the Normal_218.mp4 into normal class
"we use the model weights finetuned on UCF-Crime as in [5] to extract features. While on our TAD dataset we only use the model weights pretrained on Kinetics-400 dataset." The link to GCN https://github.com/jx-zhong-for-academic-purpose/GCN-Anomaly-Detection/tree/master/feature_extraction make me confused about it because it provided ucf-crimes -1000 and 1400 for rbg and flow ,how could I used it? If you could help me ,I'll feel so grateful!
Hello, congratulations for your work.
Were the results showed in paper 10-cropped as made by Sultani and by the RTFM? To what extent does this affect the comparison?
Hello,
Thanks for the code.
Could you please upload the model's checkpoints?
Great work!can you open source UCF-Crime pre-trained feature?
files like:
hdf5_path = "/test/UCF-Crime/UCF/gcn_feas.hdf5"
mask_path = "/test/UCF-Crime/UCF/gcn_mask.hdf5"
你好,请问TAD数据集未来会开源吗?
Your own-built dataset is about the traffic would you please show us your datasets?
Hello,
Really Appreciate your work!!. Currently I am using your repo for some my project work and I got struct in getting hdf5 formatted files.
I am UCF crime dataset and I got the extracted features from waqas sultani repo.
Can you explain me to convert the extracted features into hdf5 formatted files so that i can start training and see the results.
Thank you
Hello authors.
I request you to please add the required files that were used during the training or the format of the files that are used. For example - videos_pkl_train, hdf5 files and trained checkpoints etc.
We need to evaluate your model on our training data.
Thanks
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.