Adversarial Attacks on Deep Learning-based Video Compression and Classification Systems.
Jung-Woo ChangMojan JavaheripiSeira HidanoFarinaz KoushanfarPublished in: CoRR (2022)
Keyphrases
- classification systems
- video compression
- deep learning
- motion compensation
- video coding
- motion estimation
- unsupervised learning
- video data
- unsupervised feature learning
- motion compensated
- machine learning
- compression ratio
- discrete cosine transforms
- mental models
- weakly supervised
- learning strategies
- bit rate
- model selection
- image quality
- training set
- multiscale
- three dimensional