XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning

AAAI 2024

Pritam Sarkar   Ali Etemad

[Paper] [Code] [Website]

Updates

• Paper
• Pretrained model weights
• Evaluation codes
• Training codes
• More documentations
• Supplementary material

** Please check the project website for more details. The codes will be released soon. You may follow this repo to receive future updates. **

Abstract

We present XKD, a novel self-supervised framework to learn meaningful representations from unlabelled videos. XKD is trained with two pseudo objectives. First, masked data reconstruction is performed to learn modality-specific representations from audio and visual streams. Next, self-supervised cross-modal knowledge distillation is performed between the two modalities through a teacher-student setup to learn complementary information. We introduce a novel domain alignment strategy to tackle domain discrepancy between audio and visual modalities enabling effective cross-modal knowledge distillation. Additionally, to develop a general-purpose network capable of handling both audio and visual streams, modality-agnostic variants of XKD are introduced, which use the same pretrained backbone for different audio and visual tasks. Our proposed cross-modal knowledge distillation improves video action classification by 8% to 14% on UCF101, HMDB51, and Kinetics400. Additionally, XKD improves multimodal action classification by 5.5% on Kinetics-Sound. XKD shows state-of-the-art performance in sound classification on ESC50, achieving top-1 accuracy of 96.5%.

Citation

If you find this repository useful, please consider giving a star ⭐ and citation using the given BibTeX entry:

@misc{sarkar2022xkd,
      title={XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning}, 
      author={Pritam Sarkar and Ali Etemad},
      year={2022},
      eprint={2211.13929},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

Acknowledgments

We are grateful to Bank of Montreal and Mitacs for funding this research. We are also thankful to SciNet HPC Consortium for helping with the computation resources.

Question

You may directly contact me at pritam.sarkar@queensu.ca or connect with me on LinkedIn.