Comparative analysis of vision-based HIP models
| Models | Features |
|---|---|
| Two-stream-based CNNs (Li et al., 2020) | Human action is determined by calculating the L2 distance of the positions of the human joints between frames (CNNs) |
| CLSTM (Sarabu and Santra, 2021) | Present a two-stream network with two CNNs and Convolution Long-Short Term Memory (CLSTM) (CNN + LSTM) a powerful feature extractor in human action recognition in videos |
| CLSTDN (Saif et al., 2023) | This research proposes Convolutional Long Short-Term Deep Network (CLSTDN) consists of Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) for Recognition of human intention |
| Transformer and Bi-LSTM (Zhang et al., 2024) | extract features from human motion trajectories by analyzing changes in human joint distances with a Transformer and a Bi-LSTM, respectively |
| GRU-CNN (Du et al., 2025) | proposed multi-channel parallel GRU-CNN neural network combines the temporal analysis capabilities of GRU with the spatial feature extraction strengths of CNN through a weight allocation strategy for trajectory prediction and intention recognition |
| Models | Features |
|---|---|
| Two-stream-based CNNs ( | Human action is determined by calculating the L2 distance of the positions of the human joints between frames (CNNs) |
| CLSTM ( | Present a two-stream network with two CNNs and Convolution Long-Short Term Memory (CLSTM) (CNN + LSTM) a powerful feature extractor in human action recognition in videos |
| CLSTDN ( | This research proposes Convolutional Long Short-Term Deep Network (CLSTDN) consists of Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) for Recognition of human intention |
| Transformer and Bi-LSTM ( | extract features from human motion trajectories by analyzing changes in human joint distances with a Transformer and a Bi-LSTM, respectively |
| GRU-CNN ( | proposed multi-channel parallel GRU-CNN neural network combines the temporal analysis capabilities of GRU with the spatial feature extraction strengths of CNN through a weight allocation strategy for trajectory prediction and intention recognition |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.