virtual conference icassp 2021

A Framework For Pruning Deep Neural Networks Using Energy-Based Models

Speaker-Independent Brain Enhanced Speech Denoising

Classification Of Expert-Novice Level Using Eye Tracking And Motion Data Via Conditional Multimodal Variational Autoencoder

Estimating Fiedler Value On Large Networks Based On Random Walk Observations

Seehear: Signer Diarisation And A New Dataset

Synchronous Multi-Bit Audio Watermarking Based On Phase Shifting

Strong Data Augmentation Sanitizes Poisoning And Backdoor Attacks Without An Accuracy Tradeoff

Self-Attention Generative Adversarial Network For Speech Enhancement

Violence Detection In Videos Based On Fusing Visual And Audio Information

Enabling Efficient And Expressive Spatial Keyword Queries On Encrypted Data

Prototype-Based Personalized Pruning

Reweighted Dynamic Group Convolution