Self-supervised learning (SSL) speech models such as wav2vec and HuBERT ...
Masked auto-encoding is a popular and effective self-supervised learning...
Transformer has demonstrated promising performance in many 2D vision tas...
Domain adaptive semantic segmentation aims to learn a model with the
sup...
Label assignment (LA), which aims to assign each training sample a posit...
Current geometry-based monocular 3D object detection models can efficien...