Skeleton Based Action Recognition On Ntu Rgbd
Metrics
Accuracy (CS)
Accuracy (CV)
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Accuracy (CS) | Accuracy (CV) |
|---|---|---|
| spatial-temporal-graph-convolutional-networks-1 | 90.7 | 96.5 |
| hierarchical-recurrent-neural-network-for-3 | 59.1 | 64.0 |
| action-capsules-human-skeleton-action | 90 | 96.3 |
| actionlet-dependent-contrastive-learning-for | 84.3 | 88.8 |
| temporal-extension-module-for-skeleton-based-1 | 91.0 | 96.5 |
| a-comparative-review-of-recent-kinect-based | 83 | 89 |
| eleatt-rnn-adding-attentiveness-to-neurons-in | 80.7 | 88.4 |
| independently-recurrent-neural-network-indrnn | 81.8 | 88.0 |
| online-skeleton-based-action-recognition-with | 86.0 | 93.1 |
| online-skeleton-based-action-recognition-with | 88.3 | 95 |
| learning-multi-granular-spatio-temporal-graph | 92.0 | 96.6 |
| a-new-representation-of-skeleton-sequences | 79.6 | 84.8 |
| hypergraph-transformer-for-skeleton-based | 92.9 | 96.5 |
| learning-shape-motion-representations-from | 82.83 | 90.05 |
| non-local-graph-convolutional-networks-for-1 | 88.5 | 95.1 |
| channel-wise-topology-refinement-graph | 92.4 | 96.8 |
| spatio-temporal-lstm-with-trust-gates-for-3d | 69.2 | 77.7 |
| skeleton-based-action-recognition-with-2 | 83.2 | 89.3 |
| hyperbolic-self-paced-learning-for-self | 89.1 | 95.2 |
| online-skeleton-based-action-recognition-with | 86.3 | 92.4 |
| skeleton-based-action-recognition-with-multi | 89.4 | 96.0 |
| skeletal-quads-human-action-recognition-using | 38.6 | 41.4 |
| jointly-learning-heterogeneous-features-for-1 | 60.2 | 65.2 |
| skelemotion-a-new-representation-of-skeleton | 76.5 | 84.7 |
| multi-scale-spatial-temporal-convolutional | 92.6 | 97.8 |
| an-end-to-end-spatio-temporal-attention-model | 73.4 | 81.2 |
| an-attention-enhanced-recurrent-graph | 85.1 | 93.2 |
| part-based-graph-convolutional-network-for | 87.5 | 93.2 |
| degcn-deformable-graph-convolutional-networks | 93.6 | 97.4 |
| skeleton-based-action-recognition-with-multi | 90.0 | 96.2 |
| skeleton-based-action-recognition-with-1 | 84.8 | 92.4 |
| human-action-recognition-by-representing-3d-1 | 50.1 | 52.8 |
| online-skeleton-based-action-recognition-with | 88.9 | 94.8 |
| disentangling-and-unifying-graph-convolutions | 91.5 | 96.2 |
| motionbert-unified-pretraining-for-human | 93.0 | 97.2 |
| making-the-invisible-visible-action | 86.8 | 91.6 |
| view-adaptive-recurrent-neural-networks-for | 79.2 | 87.6 |
| global-context-aware-attention-lstm-networks | 76.10 | 84.00 |
| ensemble-deep-learning-for-skeleton-based | 74.60 | 81.25 |
| hyperbolic-self-paced-learning-for-self | 86.5 | 93.5 |
| hulk-a-universal-knowledge-translator-for | 94 | - |
| enhanced-skeleton-visualization-for-view | 80.0 | 87.2 |
| predictively-encoded-graph-convolutional | 85.6 | 93.4 |
| step-catformer-spatial-temporal-effective | 93.2 | 97.3 |
| online-skeleton-based-action-recognition-with | 86 | 93.4 |
| spatial-temporal-graph-convolutional-networks-1 | 90.1 | 95.1 |
| centrality-graph-convolutional-networks-for | 90.3 | 96.4 |
| revisiting-skeleton-based-action-recognition | 94.1 | 97.1 |
| dynamic-gcn-context-enriched-topology | 91.5 | 96.0 |
| learning-skeletal-graph-neural-networks-for | 91.6 | 96.7 |
| adaptive-rnn-tree-for-large-scale-human-1 | 74.6 | 83.2 |
| hierarchically-decomposed-graph-convolutional | 93.4 | 97.2 |
| action-recognition-for-privacy-preserving | 89.57 | 94.90 |
| an-attention-enhanced-graph-convolutional | 89.2 | 95.0 |
| generalized-graph-convolutional-networks-for | 87.5 | 94.3 |
| skeleton-based-relational-modeling-for-action | 80.7 | 88.8 |
| hulk-a-universal-knowledge-translator-for | 94.3 | - |
| self-attention-network-for-skeleton-based | 87.2 | 92.7 |
| decoupled-spatial-temporal-attention-network | 91.5 | 96.4 |
| richlt-activated-graph-convolutional-network | 85.8 | 93.0 |
| richly-activated-graph-convolutional-network | 87.3 | 93.6 |
| tensor-representations-for-action-recognition | 91.56 | 94.75 |
| ntu-rgbd-a-large-scale-dataset-for-3d-human | 62.93 | 70.27 |
| spatial-temporal-transformer-network-for | 89.9 | 96.1 |
| llms-are-good-action-recognizers | 95 | 98.4 |
| context-aware-cross-attention-for-skeleton | 84.23 | 89.27 |
| adding-attentiveness-to-the-neurons-in | 79.8 | 87.1 |
| skeleton-based-action-recognition-with | 89.1 | 94.9 |
| infogcn-representation-learning-for-human | 93.0 | 97.1 |
| investigation-of-different-skeleton-features | - | 82.31 |
| vertex-feature-encoding-and-hierarchical | 85.3 | 92.8 |
| bayesian-graph-convolution-lstm-for-skeleton | 81.8 | 89.0 |
| skeleton-based-action-recognition-using-1 | 85.0 | - |
| psumnet-unified-modality-part-streams-are-all | 92.9 | 96.7 |
| action-recognition-with-multi-stream-motion | 92.9 | 96.9 |
| constructing-stronger-and-faster-baselines | 89.9 | 94.7 |
| autogcn-towards-generic-human-activity | 88.3 | 95.5 |
| feedback-graph-convolutional-network-for | 90.2 | 96.3 |
| skateformer-skeletal-temporal-transformer-for | 93.5 | 97.8 |
| leveraging-third-order-features-in-skeleton | 91.7 | 96.4 |
| temporal-decoupling-graph-convolutional | 92.8 | 96.8 |
| pose-refinement-graph-convolutional-network | 85.2 | 91.7 |
| graph-contrastive-learning-for-skeleton-based | 93.1 | 97.0 |
| masked-motion-predictors-are-strong-3d-action | 93.1 | 97.5 |
| richlt-activated-graph-convolutional-network | 85.9 | 93.5 |
| three-stream-convolutional-neural-network | 88.6 | 93.7 |
| spatial-temporal-graph-convolutional-networks-1 | 81.5 | 88.3 |
| usdrl-unified-skeleton-based-dense | 87.1 | 93.2 |
| skeletonnet-mining-deep-part-features-for-3-d | 75.9 | 81.2 |
| spatial-residual-layer-and-dense-connection | 89.58 | 95.74 |
| constructing-stronger-and-faster-baselines | 92.1 | 96.1 |
| stronger-faster-and-more-explainable-a-graph | 90.9 | 96 |
| spatial-temporal-graph-convolutional-networks-1 | 86.6 | 93.2 |
| maskclr-attention-guided-contrastive-learning | 93.9 | 97.3 |
| dg-stgcn-dynamic-spatial-temporal-modeling | 93.2 | 97.5 |
| a-comparative-review-of-recent-kinect-based | 83.36 | 88.84 |
| a-fine-to-coarse-convolutional-neural-network | 79.6 | 84.6 |
| 3d-cnns-on-distance-matrices-for-human-action | 82.0 | 89.5 |
| spatio-temporal-lstm-with-trust-gates-for-3d | 61.70 | 75.50 |
| actional-structural-graph-convolutional | 86.8 | 94.2 |
| skeleton-based-action-recognition-via | 92.8 | 97.0 |
| pyskl-towards-good-practices-for-skeleton | 92.6 | 97.4 |
| memory-attention-networks-for-skeleton-based | 82.67 | 93.22 |
| skeleton-based-action-recognition-with-shift | 90.7 | 96.5 |
| revealing-key-details-to-see-differences-a | 93.8 | 97.8 |
| deep-independently-recurrent-neural-network | 86.70 | 93.97 |
| shap-mix-shapley-value-guided-mixing-for-long | 93.7 | 97.1 |
| mix-dimension-in-poincare-geometry-for-3d | 89.7 | 96 |
| ntu-rgbd-a-large-scale-dataset-for-3d-human | 60.7 | 67.3 |
| semantics-guided-neural-networks-for | 89.0 | 94.5 |
| skeleton-based-action-recognition-with-4 | 89.9 | 96.1 |
| joint-mixing-data-augmentation-for-skeleton | 93.7 | 97.2 |
| symbiotic-graph-neural-networks-for-3d | 90.1 | 96.4 |
| blockgcn-redefine-topology-awareness-for | 93.1 | 97.0 |
| a-semantics-guided-graph-convolutional | 86.2 | 94.2 |
| constructing-stronger-and-faster-baselines | 90.9 | 95.5 |
| learning-human-pose-models-from-synthesized | 80.9 | 86.1 |
| modeling-temporal-dynamics-and-spatial | 71.3 | 79.5 |
| online-skeleton-based-action-recognition-with | 84.1 | 92.6 |
| skeleton-image-representation-for-3d-action | 73.3 | 80.3 |
| msa-gcn-exploiting-multi-scale-temporal | 93.6 | 97.4 |
| co-occurrence-feature-learning-from-skeleton | 86.5 | 91.1 |
| online-skeleton-based-action-recognition-with | 86.3 | 93.8 |
| language-knowledge-assisted-representation | 93.5 | 97.2 |
| view-adaptive-neural-networks-for-high | 89.4 | 95.0 |
| spatial-temporal-graph-attention-network-for | 92.8 | 97.3 |
| multi-modality-co-learning-for-efficient-1 | 93.5 | 97.4 |
| focusing-and-diffusion-bidirectional | 90.3 | 96.3 |
| pyskl-towards-good-practices-for-skeleton | 91.4 | 98.3 |
| pgcn-tca-pseudo-graph-convolutional-network | 88.0 | 93.6 |
| language-supervised-training-for-skeleton | 92.9 | 97 |
| interpretable-3d-human-action-analysis-with | 74.3 | 83.1 |
| learning-graph-convolutional-network-for | 89.4 | 95.7 |