Differentiable Frequency-based Disentanglement for Aerial Video Action Recognition
This work addresses the challenge of recognizing human actions in aerial videos with small actors and dynamic backgrounds, offering incremental improvements for UAV-based surveillance or monitoring applications.
The paper tackles human activity recognition in UAV videos by proposing a learning algorithm that uses a differentiable frequency mask prior and a cost-function for frame sampling, achieving relative improvements of 5.72% to 13.00% over state-of-the-art methods.
We present a learning algorithm for human activity recognition in videos. Our approach is designed for UAV videos, which are mainly acquired from obliquely placed dynamic cameras that contain a human actor along with background motion. Typically, the human actors occupy less than one-tenth of the spatial resolution. Our approach simultaneously harnesses the benefits of frequency domain representations, a classical analysis tool in signal processing, and data driven neural networks. We build a differentiable static-dynamic frequency mask prior to model the salient static and dynamic pixels in the video, crucial for the underlying task of action recognition. We use this differentiable mask prior to enable the neural network to intrinsically learn disentangled feature representations via an identity loss function. Our formulation empowers the network to inherently compute disentangled salient features within its layers. Further, we propose a cost-function encapsulating temporal relevance and spatial content to sample the most important frame within uniformly spaced video segments. We conduct extensive experiments on the UAV Human dataset and the NEC Drone dataset and demonstrate relative improvements of 5.72% - 13.00% over the state-of-the-art and 14.28% - 38.05% over the corresponding baseline model.