CoChat

Visual Forecasting by Imitating Dynamics in Natural Sequences

Open access
Zeng, Kuo-Hao, Shen, William B., Huang, De-An, Sun, Min, Niebles, Juan Carlos · 2017 · 2017 IEEE International Conference on Computer Vision (ICCV)
Cited by 53

The DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.

Abstract

We introduce a general framework for visual forecasting, which directly imitates visual sequences without additional supervision. As a result, our model can be applied at several semantic levels and does not require any domain knowledge or handcrafted features. We achieve this by formulating visual forecasting as an inverse reinforcement learning (IRL) problem, and directly imitate the dynamics in natural sequences from their raw pixel values. The key challenge is the high-dimensional and continuous state-action space that prohibits the application of previous IRL algorithms. We address this computational bottleneck by extending recent progress in model-free imitation with trainable deep feature representations, which (1) bypasses the exhaustive state-action pair visits in dynamic programming by using a dual formulation and (2) avoids explicit state sampling at gradient computation using a deep feature reparametrization. This allows us to apply IRL at scale and directly imitate the dynamics in high-dimensional continuous visual sequences from the raw pixel values. We evaluate our approach at three different level-of-abstraction, from low level pixels to higher level semantics: future frame generation, action anticipation, visual story forecasting. At all levels, our approach outperforms existing methods.

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/iccv.2017.326

Cites

15 of 15 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: 39

  1. soomro, 2012, Ucf101 A Dataset of 101 Human Actions Classes from Videos in the Wild
  2. sigurdsson, 2016, Learning visual storylines with skipping recurrent neural networks, ECCV
  3. 10.1109/CVPR.2016.508 · DOI · Paper page
  4. 10.1145/1102351.1102433 · DOI
  5. mnih, 2014, Recurrent models of visual attention, NIPS
  6. radford, 2016, Unsupervised representation learning with deep convolutional generative adversarial networks, ICLRE
  7. patron-perez, 2010, High five: Recognising human interactions in tv shows, BMVC
  8. pathak, 2017, Learning features by watching objects move, CVPR
  9. mathieu, 2016, Deep multiscale video prediction beyond mean square error, ICLRE
  10. mathe, 2013, Action from still image dataset and inverse optimal control to learn task specific visual scanpaths, NIPS
  11. 10.1109/CDC.2014.7040156 · DOI
  12. 10.1109/CVPR.2016.110 · DOI · Paper page
  13. 10.1177/0278364915619772 · DOI · Paper page
  14. lan, 2014, A hierarchical representation for future action prediction, ECCV
  15. ma, 2017, Forecasting interactive dynamics of pedestrians with fictitious play, CVPR
  16. li, 2017, Adversarial learning for neural dialogue generation, EMNLP
  17. xue, 2016, Visual dynamics: Probabilistic future frame synthesis via cross convolutional networks, NIPS
  18. yan, 2016, Attribute2image: Conditional image generation from visual attributes, ECCV
  19. ziebart, 2008, Maximum entropy inverse reinforcement learning, AAAI
  20. zeng, 2017, Agent-centric risk assessment: Accident anticipation and risky region localization, CVPR
  21. yu, 2017, Seqgan: sequence generative adversarial nets with policy gradient, AAAI, 10.1609/aaai.v31i1.10804 · DOI · Paper page
  22. 10.1109/ICCV.2015.530 · DOI
  23. ho, 2016, Model-free imitation learning with policy optimization, ICML
  24. huang, 2014, Action-reaction: Forecasting the dynamics of human interaction, ECCV
  25. 10.18653/v1/N16-1147 · DOI · Paper page
  26. isola, 2017, Image-to-image translation with conditional adversarial networks, CVPR
  27. kalchbrenner, 2016, Video pixel networks
  28. chan, 2016, Anticipating accidents in dashcam videos, ACCV
  29. 10.1109/CVPR.2009.5206540 · DOI
  30. finn, 2016, Guided cost learning: Deep inverse optimal control via policy optimization, ICML
  31. gorban, 2015, THUMOS Challenge Action Recognition with A Large Number of Classes
  32. goodfellow, 2014, Generative adversarial nets, NIPS
  33. xie, 2013, Inferring “dark matter” and “dark energy” from videos, ICCV
  34. 10.1109/ICRA.2017.7989385 · DOI · Paper page
  35. 10.1109/CVPR.2014.416 · DOI · Paper page
  36. walker, 2016, An uncertain future: Forecasting from variational autoencoders, ECCV
  37. 10.1109/ICCV.2015.281 · DOI · Paper page
  38. 10.1109/CVPR.2016.18 · DOI · Paper page
  39. vondrick, 2017, Generating the future with adversarial transformers, CVPR

Cited by

See all 53

Metadata

Authors
Zeng, K.-H., Shen, W. B., Huang, D.-A., Sun, M., & Niebles, J. C.
Venue
2017 IEEE International Conference on Computer Vision (ICCV)
Publisher
IEEE
Published
2017-10
Pages
3018-3027
OpenAlex
W2963302737
Cited by
53
References
15
FWCI
2.10
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.