Reconstruction of Fragmented Trajectories of Collective Motion using Hadamard Deep Autoencoders
This work addresses trajectory fragmentation issues in multi-object tracking for collective motion analysis, but it appears incremental as it extends existing deep autoencoder methods with a specific loss function.
The paper tackles the problem of reconstructing fragmented trajectories in collective motion data, such as from fish or humans, by introducing a Hadamard deep autoencoder (HDA) trained only on fully observed segments, and it compares performance with a low-rank matrix completion scheme.
Learning dynamics of collectively moving agents such as fish or humans is an active field in research. Due to natural phenomena such as occlusion and change of illumination, the multi-object methods tracking such dynamics might lose track of the agents where that might result fragmentation in the constructed trajectories. Here, we present an extended deep autoencoder (DA) that we train only on fully observed segments of the trajectories by defining its loss function as the Hadamard product of a binary indicator matrix with the absolute difference between the outputs and the labels. The trajectories of the agents practicing collective motion is low-rank due to mutual interactions and dependencies between the agents that we utilize as the underlying pattern that our Hadamard deep autoencoder (HDA) codes during its training. The performance of our HDA is compared with that of a low-rank matrix completion scheme in the context of fragmented trajectory reconstruction.