|TREND||DATASET||BEST METHOD||PAPER TITLE||PAPER||CODE||COMPARE|
We analyze key properties of the approach that make it work, finding that the contrastive loss outperforms a popular alternative based on cross-view prediction, and that the more views we learn from, the better the resulting representation captures underlying scene semantics.
#6 best model for Self-Supervised Image Classification on ImageNet
The objective of this paper is self-supervised learning of spatio-temporal embeddings from video, suitable for human action recognition.
SOTA for Self-Supervised Action Recognition on UCF101 (using extra training data)
We conduct extensive experiments with C3D to validate the effectiveness of our proposed approach.
#12 best model for Self-Supervised Action Recognition on UCF101
We evaluate our self-supervised trained TCE model by adding a classification layer and finetuning the learned representation on the downstream task of video action recognition on the UCF101 dataset.
#3 best model for Self-Supervised Action Recognition on UCF101