Action Recognition

Benchmarks

TREND DATASET BEST METHOD PAPER TITLE PAPER CODE COMPARE

Latest papers with code

Play Fair: Frame Attributions in Video Models

24 Nov 2020willprice/play-fair

We offer detailed analysis of supporting/distracting frames, and the relationships of ESVs to the frame's position, class prediction, and sequence length.

ACTION RECOGNITION RELATIONAL REASONING

9
24 Nov 2020

Prototypical Contrast and Reverse Prediction: Unsupervised Skeleton Based Action Recognition

14 Nov 2020Mikexu007/PCRP

Existing approaches usually learn action representations by sequential prediction but they suffer from the inability to fully learn semantic information.

ACTION RECOGNITION SEMANTIC SIMILARITY SEMANTIC TEXTUAL SIMILARITY SKELETON BASED ACTION RECOGNITION UNSUPERVISED REPRESENTATION LEARNING

1
14 Nov 2020

CNN based Multistage Gated Average Fusion (MGAF) for Human Action Recognition Using Depth and Inertial Sensors

29 Oct 2020zaamad/Gated-Average-Fusion

Experiments on three publicly available multimodal HAR datasets demonstrate that the proposed MGAF outperforms the previous state of the art fusion methods for depth-inertial HAR in terms of recognition accuracy while being computationally much more efficient.

ACTION RECOGNITION

0
29 Oct 2020

View-Invariant, Occlusion-Robust Probabilistic Embedding for Human Pose

23 Oct 2020google-research/google-research

We further show that keypoint occlusion augmentation during training significantly improves retrieval performance on partial 2D input poses.

3D POSE ESTIMATION ACTION RECOGNITION VIDEO ALIGNMENT

14,451
23 Oct 2020

Deep Analysis of CNN-based Spatio-temporal Representations for Action Recognition

22 Oct 2020IBM/action-recognition-pytorch

In recent years, a number of approaches based on 2D CNNs and 3D CNNs have emerged for video action recognition, achieving state-of-the-art results on several large-scale benchmark datasets.

ACTION RECOGNITION

39
22 Oct 2020

Stronger, Faster and More Explainable: A Graph Convolutional Baseline for Skeleton-based Action Recognition

20 Oct 2020yfsong0709/ResGCNv1

However, the complexity of the State-Of-The-Art (SOTA) models of this task tends to be exceedingly sophisticated and over-parameterized, where the low efficiency in model training and inference has obstructed the development in the field, especially for large-scale action datasets.

ACTION RECOGNITION SKELETON BASED ACTION RECOGNITION

18
20 Oct 2020