Evolving Losses for Unsupervised Video Representation Learning
We present a new method to learn video representations from large-scale unlabeled video data. Ideally, this representation will be generic and transferable, directly usable for new tasks such as action recognition and zero or few-shot learning. We formulate unsupervised representation learning as a multi-modal, multi-task learning problem, where the representations are shared across different modalities via distillation. Further, we introduce the concept of loss function evolution by using an evolutionary search algorithm to automatically find optimal combination of loss functions capturing many (self-supervised) tasks and modalities. Thirdly, we propose an unsupervised representation evaluation metric using distribution matching to a large unlabeled dataset as a prior constraint, based on Zipf's law. This unsupervised constraint, which is not guided by any labeling, produces similar results to weakly-supervised, task-specific ones. The proposed unsupervised representation learning results in a single RGB network and outperforms previous methods. Notably, it is also more effective than several label-based methods (e.g., ImageNet), with the exception of large, fully labeled video datasets.
Code (0)
등록된 구현이 없습니다.
Tasks
Action RecognitionFew-Shot LearningMulti-Task LearningRepresentation LearningSelf-Supervised Action RecognitionSimilar Papers 제목 키워드 기반
Evolving Losses for Unlabeled Video Representation Learning
We present a new method to learn video representations from unlabeled data. Given large-scale unlabeled video data, the objective is to benefit from such data by learning a generic and transferable representation space t…
Few-Shot LearningMulti-Task LearningOptical Flow EstimationRepresentation LearningContrastive Losses Are Natural Criteria for Unsupervised Video Summarization
Video summarization aims to select the most informative subset of frames in a video to facilitate efficient video browsing. Unsupervised methods usually rely on heuristic training objectives such as diversity and represe…
Diversityimage-classificationImage ClassificationRepresentation Learning+2Entropy-driven Unsupervised Keypoint Representation Learning in Videos
Extracting informative representations from videos is fundamental for effectively learning various downstream tasks. We present a novel approach for unsupervised learning of meaningful representations from videos, levera…
Representation LearningDeep Association Learning for Unsupervised Video Person Re-identification
Deep learning methods have started to dominate the research progress of video-based person re-identification (re-id). However, existing methods mostly consider supervised learning, which requires exhaustive manual effort…
Person Re-IdentificationUnsupervised Person Re-IdentificationVideo-Based Person Re-IdentificationSlotMatch: Distilling Object-Centric Representations for Unsupervised Video Segmentation
Unsupervised video segmentation is a challenging computer vision task, especially due to the lack of supervisory signals coupled with the complexity of visual scenes. To overcome this challenge, state-of-the-art models b…
Knowledge DistillationVideo Segmentation