Spatiotemporal Contrastive Video Representation Learning
Research output: Contribution to journal › Conference article › Research › peer-review
We present a self-supervised Contrastive Video Representation Learning (CVRL) method to learn spatiotemporal visual representations from unlabeled videos. Our representations are learned using a contrastive loss, where two augmented clips from the same short video are pulled together in the embedding space, while clips from different videos are pushed away. We study what makes for good data augmentations for video self-supervised learning and find that both spatial and temporal information are crucial. We carefully design data augmentations involving spatial and temporal cues. Concretely, we propose a temporally consistent spatial augmentation method to impose strong spatial augmentations on each frame of the video while maintaining the temporal consistency across frames. We also propose a sampling-based temporal augmentation method to avoid overly enforcing invariance on clips that are distant in time. On Kinetics-600, a linear classifier trained on the representations learned by CVRL achieves 70.4% top-1 accuracy with a 3D-ResNet-50 (R3D-50) backbone, outperforming ImageNet supervised pre-training by 15.7% and SimCLR unsupervised pre-training by 18.8% using the same inflated R3D-50. The performance of CVRL can be further improved to 72.9% with a larger R3D-152 (2× filters) backbone, significantly closing the gap between unsupervised and supervised video representation learning. Our code and models will be available at https://github.com/tensorflow/models/tree/master/official/.
Original language | English |
---|---|
Journal | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition |
Pages (from-to) | 6960-6970 |
Number of pages | 11 |
ISSN | 1063-6919 |
DOIs | |
Publication status | Published - 2021 |
Externally published | Yes |
Event | 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021 - Virtual, Online, United States Duration: 19 Jun 2021 → 25 Jun 2021 |
Conference
Conference | 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021 |
---|---|
Country | United States |
City | Virtual, Online |
Period | 19/06/2021 → 25/06/2021 |
Bibliographical note
Funding Information:
We would like to thank Yeqing Li and the TensorFlow TPU team for their infrastructure support; Tsung-Yi Lin, Ting Chen and Yonglong Tian for their valuable feedback.
Publisher Copyright:
© 2021 IEEE
ID: 301817502