Temporal Gaussian Mixture Layer for Videos
This addresses the challenge of efficient temporal modeling in video analysis for applications like activity detection, though it appears incremental as it builds on existing convolutional frameworks.
The paper tackles the problem of capturing longer-term temporal information in continuous activity videos by introducing a Temporal Gaussian Mixture (TGM) layer, resulting in significant performance improvements over state-of-the-art methods on datasets like Charades and MultiTHUMOS.
We introduce a new convolutional layer named the Temporal Gaussian Mixture (TGM) layer and present how it can be used to efficiently capture longer-term temporal information in continuous activity videos. The TGM layer is a temporal convolutional layer governed by a much smaller set of parameters (e.g., location/variance of Gaussians) that are fully differentiable. We present our fully convolutional video models with multiple TGM layers for activity detection. The extensive experiments on multiple datasets, including Charades and MultiTHUMOS, confirm the effectiveness of TGM layers, significantly outperforming the state-of-the-arts.