Reinforced Temporal Attention and Split-Rate Transfer for Depth-Based Person Re-Identification
This addresses person re-identification for security/surveillance using depth data, with incremental improvements in transfer learning and temporal modeling.
The paper tackles person re-identification from depth sensors by introducing split-rate RGB-to-depth transfer to address data scarcity and reinforced temporal attention for video sequences, achieving large performance gains over state-of-the-art RGB models in scenarios with unseen clothing.
We address the problem of person re-identification from commodity depth sensors. One challenge for depth-based recognition is data scarcity. Our first contribution addresses this problem by introducing split-rate RGB-to-Depth transfer, which leverages large RGB datasets more effectively than popular fine-tuning approaches. Our transfer scheme is based on the observation that the model parameters at the bottom layers of a deep convolutional neural network can be directly shared between RGB and depth data while the remaining layers need to be fine-tuned rapidly. Our second contribution enhances re-identification for video by implementing temporal attention as a Bernoulli-Sigmoid unit acting upon frame-level features. Since this unit is stochastic, the temporal attention parameters are trained using reinforcement learning. Extensive experiments validate the accuracy of our method in person re-identification from depth sequences. Finally, in a scenario where subjects wear unseen clothes, we show large performance gains compared to a state-of-the-art model which relies on RGB data.