Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Cited by 231

The DOI matches its Crossref record; 20 of the 20 references it cites were found in a registry or an index.

VideoMoCo: Contrastive Video Representation Learning with Temporally Adversarial Examples

Open access
Pan, Tian, Song, Yibing, Yang, Tianyu, Jiang, Wenhao, Liu, Wei · 2021 · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Abstract

MoCo [11] is effective for unsupervised image representation learning. In this paper, we propose VideoMoCo for unsupervised video representation learning. Given a video sequence as an input sample, we improve the temporal feature representations of MoCo from two perspectives. First, we introduce a generator to drop out several frames from this sample temporally. The discriminator is then learned to encode similar feature representations regardless of frame removals. By adaptively dropping out different frames during training iterations of adversarial learning, we augment this input sample to train a tempo-rally robust encoder. Second, we use temporal decay to model key attenuation in the memory queue when computing the contrastive loss. As the momentum encoder updates after keys enqueue, the representation ability of these keys degrades when we use the current input sample for contrastive learning. This degradation is reflected via temporal decay to attend the input sample to recent keys in the queue. As a result, we adapt MoCo to learn video representations without empirically designing pretext tasks. By empowering the temporal robustness of the encoder and modeling the temporal decay of the keys, our VideoMoCo improves MoCo temporally based on contrastive learning. Experiments on benchmark datasets including UCF101 and HMDB51 show that VideoMoCo stands as a state-of-the-art video representation learning method.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/cvpr46437.2021.01105

Cites

20 of 20 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 32

  1. 10.1109/ICCV.2019.00756 · DOI · Paper page
  2. 10.1109/ICCV.2019.00810 · DOI
  3. 10.1109/ICCV.2017.606 · DOI · Paper page
  4. soomro, 2012, A dataset of 101 human action classes from videos in the wild, Computer and Vision Research Center
  5. 10.1109/CVPR.2018.00937 · DOI · Paper page
  6. 10.1109/ICCV.2017.279 · DOI · Paper page
  7. arjovsky, 2017, Wasserstein generative adversarial networks, International Conference on Machine Learning
  8. chen, 2020, A simple framework for contrastive learning of visual representations, International Conference on Machine Learning
  9. buchler, 2018, Improving spatiotemporal self-supervision by deep reinforcement learning, European Conference on Computer Vision
  10. korbar, 2018, Cooperative learning of audio and video models from self-supervised synchronization, Neural Information Processing Systems
  11. 10.1109/2.36 · DOI · Paper page
  12. 10.1109/ICCV.2017.79 · DOI · Paper page
  13. 10.1609/aaai.v34i07.6840 · DOI
  14. liu, 2020, Rethinking image inpainting via a mutual encoder-decoder with feature equalizations, European Conference on Computer Vision
  15. 10.1109/CVPR.2019.01058 · DOI · Paper page
  16. 10.1109/CVPR42600.2020.00672 · DOI · Paper page
  17. 10.1109/CVPR42600.2020.00658 · DOI · Paper page
  18. 10.1109/ICCVW.2019.00186 · DOI · Paper page
  19. tian, 2020, Con-trastive multiview coding, European Conference on Computer Vision
  20. jenni, 2020, Video representation learning by recognizing temporal transformations, European Conference on Computer Vision
  21. jia, 2020, Robust tracking against adversarial attacks, European Conference on Computer Vision
  22. 10.1609/aaai.v33i01.33018545 · DOI · Paper page
  23. 10.1109/CVPR.2019.00190 · DOI · Paper page
  24. goodfellow, 2014, Generative adversarial nets, Neural Information Processing Systems
  25. 10.1109/CVPR.2018.00586 · DOI
  26. 10.1109/CVPR.2017.525 · DOI · Paper page
  27. 10.1109/CVPR.2019.00267 · DOI · Paper page
  28. 10.1007/s11263-020-01357-4 · DOI
  29. wang, 2020, Rethinking image deraining via rain streaks and vapors, European Conference on Computer Vision
  30. vondrick, 2018, Tracking emerges by colorizing videos, European Conference on Computer Vision
  31. 10.1109/CVPR.2019.00140 · DOI · Paper page
  32. wang, 2020, Self-supervised video representation learning by pace prediction, European Conference on Computer Vision

Cited by

See all 231

Metadata

Authors
Pan, T., Song, Y., Yang, T., Jiang, W., & Liu, W.
Venue
2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2021-06
Pages
11200-11209
ISSN
1063-6919 (ISSN-L)
OpenAlex
W3176780013
Cited by
231
References
20
FWCI
16.46
Type
conference paper
Language
en
Open access
green

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.