PreViTS: Contrastive Pretraining with Video Tracking Supervision
This work addresses the challenge of learning robust visual representations from videos for tasks like action classification, though it appears incremental as it builds on existing contrastive methods.
The paper tackles the problem of imperfect supervisory signals in self-supervised learning from videos by proposing PreViTS, which uses unsupervised tracking to select clips with the same object and spatially constrain learning, resulting in state-of-the-art performance on action classification.
Videos are a rich source for self-supervised learning (SSL) of visual representations due to the presence of natural temporal transformations of objects. However, current methods typically randomly sample video clips for learning, which results in an imperfect supervisory signal. In this work, we propose PreViTS, an SSL framework that utilizes an unsupervised tracking signal for selecting clips containing the same object, which helps better utilize temporal transformations of objects. PreViTS further uses the tracking signal to spatially constrain the frame regions to learn from and trains the model to locate meaningful objects by providing supervision on Grad-CAM attention maps. To evaluate our approach, we train a momentum contrastive (MoCo) encoder on VGG-Sound and Kinetics-400 datasets with PreViTS. Training with PreViTS outperforms representations learnt by contrastive strategy alone on video downstream tasks, obtaining state-of-the-art performance on action classification. PreViTS helps learn feature representations that are more robust to changes in background and context, as seen by experiments on datasets with background changes. Learning from large-scale videos with PreViTS could lead to more accurate and robust visual feature representations.