Intelligent home interactive gesture recognition method based on multimodal feature fusion and DTW algorithm
Yan Luo & Zhaosheng Yang
What the paper says
In order to improve the accuracy of smart home interactive gesture recognition, a smart home interactive gesture recognition method based on multimodal feature fusion and DTW algorithm is proposed. The sampled gesture data are subjected to moving average filtering, and the effective segments are extracted. A CR3D network model is constructed by integrating residual connections into the 3D-CNN framework and multimodal feature fusion is performed on the video frame sequence and the video optical flow frame sequence. Subsequently, the optimal warping path is determined using the DTW algorithm. The Euclidean distance is replaced by the cumulative distance to derive the gesture recognition result. The results show that the error rate is maintained below 0.5%, the coefficient of variation for the same gesture recognised across multiple users is 0 and the minimum recognition rate for multiple gesture categories reaches 96%. These findings indicate that the proposed method achieves high recognition accuracy.
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.50 × 0.4 = 0.20 |
| M · momentum | 0.50 × 0.15 = 0.07 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.