Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/VPN (RGB + Pose)

VPN (RGB + Pose)

Reported on 19 benchmarks across 8 tasks · 1 paper · 17 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision6 results

VideoonN-UCLA
Accuracy· uses extra data· 2020-07-06
93.5
best: 99.1 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
VideoonToyota Smarthome dataset
CS· uses extra data· 2020-07-06
60.8
best: 72.9 (π-ViT)
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
VideoonToyota Smarthome dataset
CV1· uses extra data· 2020-07-06
43.8
best: 55.2 (π-ViT)
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
VideoonToyota Smarthome dataset
CV2· uses extra data· 2020-07-06
53.5
best: 67.8 (EPAM-Net)
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Temporal Action LocalizationonN-UCLA
Accuracy· uses extra data· 2020-07-06
93.5
best: 99.1 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Action LocalizationonN-UCLA
Accuracy· uses extra data· 2020-07-06
93.5
best: 99.1 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056

Time Series6 results

Action DetectiononN-UCLA
Accuracy· uses extra data· 2020-07-06
93.5
best: 99.1 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Action RecognitiononNTU RGB+D
Accuracy (CS)· uses extra data· 2020-07-06
95.5
best: 97.4 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Action RecognitiononNTU RGB+D
Accuracy (CV)· uses extra data· 2020-07-06
98
best: 99.6 (PoseC3D (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Action RecognitiononNTU RGB+D 120
Accuracy (Cross-Subject)· uses extra data· 2020-07-06
87.8
best: 95.6 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Action RecognitiononN-UCLA
Accuracy· uses extra data· 2020-07-06
93.5
best: 99.1 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Action RecognitiononNTU RGB+D 120
Accuracy (Cross-Setup)· uses extra data· 2020-07-06
86.3
best: 96.7 (DSCNet (RGB + Pose))
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056

Robots5 results

Activity RecognitiononNTU RGB+D
Accuracy (CS)· uses extra data· 2020-07-06
95.5
best: 97.4 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Activity RecognitiononNTU RGB+D
Accuracy (CV)· uses extra data· 2020-07-06
98
best: 99.6 (PoseC3D (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Activity RecognitiononNTU RGB+D 120
Accuracy (Cross-Subject)· uses extra data· 2020-07-06
87.8
best: 95.6 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Activity RecognitiononN-UCLA
Accuracy· uses extra data· 2020-07-06
93.5
best: 99.1 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056
Activity RecognitiononNTU RGB+D 120
Accuracy (Cross-Setup)· uses extra data· 2020-07-06
86.3
best: 96.7 (DSCNet (RGB + Pose))
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056

Methodology1 result

Zero-Shot LearningonN-UCLA
Accuracy· uses extra data· 2020-07-06
93.5
best: 99.1 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056

Natural Language Processing1 result

3D Action RecognitiononN-UCLA
Accuracy· uses extra data· 2020-07-06
93.5
best: 99.1 (DSCNet (RGB + Pose))
SOTA
VPN: Learning Video-Pose Embedding for Activities of Daily Living arXiv:2007.03056