Filtering Point Targets via Online Learning of Motion Models
This addresses the challenge of accurate multi-target tracking in noisy environments for applications like surveillance or sensor data processing, though it is incremental as it builds on existing filtering and learning techniques.
The paper tackles the problem of filtering point targets in cluttered and noisy data by proposing an algorithm that learns motion models online using a recurrent neural network with LSTM, combined with a novel low-complexity data association method, achieving remarkable performance in synthetic and real scenarios.
Filtering point targets in highly cluttered and noisy data frames can be very challenging, especially for complex target motions. Fixed motion models can fail to provide accurate predictions, while learning based algorithm can be difficult to design (due to the variable number of targets), slow to train and dependent on separate train/test steps. To address these issues, this paper proposes a multi-target filtering algorithm which learns the motion models, on the fly, using a recurrent neural network with a long short-term memory architecture, as a regression block. The target state predictions are then corrected using a novel data association algorithm, with a low computational complexity. The proposed algorithm is evaluated over synthetic and real point target filtering scenarios, demonstrating a remarkable performance over highly cluttered data sequences.