CoChat

Jointly Learning Energy Expenditures and Activities Using Egocentric Multimodal Signals

Open access
Nakamura, Katsuyuki, Yeung, Serena, Alahi, Alexandre, Fei-Fei, Li · 2017 · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Cited by 67

The DOI matches its Crossref record; 11 of the 11 references it cites were found in a registry or an index.

Abstract

Physiological signals such as heart rate can provide valuable information about an individuals state and activity. However, existing work on computer vision has not yet explored leveraging these signals to enhance egocentric video understanding. In this work, we propose a model for reasoning on multimodal data to jointly predict activities and energy expenditures. We use heart rate signals as privileged self-supervision to derive energy expenditure in a training stage. A multitask objective is used to jointly optimize the two tasks. Additionally, we introduce a dataset that contains 31 hours of egocentric video augmented with heart rate and acceleration signals. This study can lead to new applications such as a visual calorie counter.

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/cvpr.2017.721

Cites

11 of 11 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: 57

  1. 10.1249/MSS.0b013e31825e825a · DOI · Paper page
  2. 10.1109/CVPR.2015.7298625 · DOI · Paper page
  3. 10.1109/CVPR.2011.5995406 · DOI · Paper page
  4. 10.1109/TITB.2010.2051955 · DOI · Paper page
  5. 10.1109/CVPR.2014.223 · DOI · Paper page
  6. 10.1109/JPROC.2012.2200554 · DOI
  7. 10.1109/ICCV.2013.399 · DOI · Paper page
  8. lee, 2012, Discovering important people and objects for egocentric video summarization, CVPR
  9. 10.1109/TSMCC.2009.2023380 · DOI · Paper page
  10. 10.1038/sj.ejcn.1602264 · DOI
  11. 10.1109/CVPR.2015.7298594 · DOI · Paper page
  12. 10.1109/ICCV.2015.510 · DOI · Paper page
  13. tokui, 2015, Chainer: a next-generation open source framework for deep learning, Workshop on Machine Learning Systems at NIPS
  14. 10.1109/TCSVT.2008.2005594 · DOI · Paper page
  15. hiilloskorpi, 2003, Use of heart rate to predict energy expenditure from low to high activity levels, International Journal of Sports Medicine, 24, 332, 10.1055/s-2003-40701 · DOI · Paper page
  16. 10.4108/phat.1.1.e2 · DOI
  17. 10.1016/j.patrec.2008.08.002 · DOI · Paper page
  18. yeung, 2015, Every moment counts Dense detailed labeling of actions in complex videos
  19. idrees, 2016, The THUMOS challenge on action recognition for videos “in the wild”
  20. 10.1109/CVPR.2016.288 · DOI
  21. 10.1145/1922649.1922653 · DOI · Paper page
  22. abdullah, 2012, Towards population scale activity recognition: a framework for handling data diversity, AAAI
  23. 10.1006/cviu.1998.0716 · DOI · Paper page
  24. 10.1109/CVPR.2016.138 · DOI · Paper page
  25. hamed, 2012, Detecting activities of daily living in first-person camera views, CVPR
  26. 10.4108/icst.mobihealth.2014.257219 · DOI
  27. 10.1109/CVPR.2014.325 · DOI
  28. 10.1109/CVPRW.2009.5204354 · DOI · Paper page
  29. soomro, 2012, Ucf101 A Dataset of 101 Human Actions Classes from Videos in the Wild
  30. 10.1016/j.patcog.2016.07.031 · DOI
  31. 10.1109/CVPR.2015.7298691 · DOI
  32. 10.1109/CVPR.2013.352 · DOI · Paper page
  33. 10.1109/CVPR.2010.5540074 · DOI · Paper page
  34. 10.1079/BJN19890107 · DOI · Paper page
  35. 10.1152/japplphysiol.00818.2005 · DOI · Paper page
  36. liu, 2012, Multisensor data fusion for physical activity assessment, IEEE Trans Biomedical Engineering, 59, 687, 10.1109/TBME.2011.2178070 · DOI · Paper page
  37. damen, 2014, You-do, I-learn: discovering task relevant objects and their modes of interaction from multiuser egocentric video, BMVC
  38. dauphin, 2015, RMSProp and equilibrated adaptive learning rates for non-convex optimization
  39. 10.1109/CVPR.2015.7298878 · DOI · Paper page
  40. 10.1088/0967-3334/35/11/2191 · DOI · Paper page
  41. 10.1109/ICCV.2011.6126269 · DOI · Paper page
  42. 10.1109/CVPR.2012.6247805 · DOI · Paper page
  43. fathi, 2012, Learning to recognize daily actions using gaze, ECCV
  44. 10.1109/ICCV.2015.376 · DOI
  45. 10.1109/ICCV.2015.226 · DOI · Paper page
  46. bambach, 2015, A survey on recent advances of computer vision algorithms for egocentric video
  47. 10.1109/TCSVT.2015.2409731 · DOI · Paper page
  48. 10.1109/CVPR.2016.338 · DOI
  49. 10.1249/MSS.0b013e3181a24536 · DOI · Paper page
  50. mozer, 1989, A focused backpropagation algorithm for temporal pattern recognition, Complex Systems, 3, 349
  51. 10.1109/BSN.2013.6575520 · DOI
  52. 10.3390/s16010072 · DOI
  53. 10.1109/EMBC.2015.7318411 · DOI
  54. 10.1109/CVPR.2013.350 · DOI · Paper page
  55. 10.1016/j.cmpb.2016.02.009 · doi:10.1016/j.cmpb.2016.02.009
  56. 10.1007/978-3-642-12654-3_15 · DOI
  57. 10.1109/CVPR.2016.209 · DOI · Paper page

Cited by

See all 67

Metadata

Authors
Nakamura, K., Yeung, S., Alahi, A., & Fei-Fei, L.
Venue
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2017-07
Pages
6817-6826
ISSN
1063-6919 (ISSN-L)
OpenAlex
W2737602873
Cited by
67
References
11
FWCI
1.70
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.