ActNetFormer: Transformer-ResNet Hybrid Method for Semi-Supervised Action Recognition in Videos

2 code implementations • 9 Apr 2024 • Sharana Dharshikgan Suresh Dass, Hrishav Bakul Barua, Ganesh Krishnasamy, Raveendran Paramesran, Raphael C. -W. Phan

Our framework leverages both labeled and unlabelled data to robustly learn action representations in videos, combining pseudo-labeling with contrastive learning for effective learning from both types of samples.

Action Recognition In Videos Activity Recognition In Videos +4