Exploring Convolutional Recurrent architectures for anomaly detection in videos: a comparative study

Convolutional Recurrent architectures are currently preferred for spatio-temporal learning tasks in videos to the 3D convolutional networks which accompany a huge computational burden and it is imperative to understand the working of different architectural configurations. But most of the current works on visual learning, especially for video anomaly detection, predominantly employ ConvLSTM networks and focus less on other possible variants of Convolutional Recurrent configurations for temporal learning which warrants a need to study the different possible variants to make informed, optimal design choices according to the nature of the application at hand. We explore a variety of Convolutional Recurrent architectures and the influence of hyper-parameters on their performance for the task of anomaly detection. Through this work, we also intend to quantify the efficiency of the architectures based on the trade-off between their performance and computational complexity. With comprehensive quantitative and visual evidence, we establish that the ConvGRU based configurations are the most effective and perform better than the popular ConvLSTM configurations on video anomaly detection tasks, in contrast to what is seen from the literature.

Improving Action Recognition via Temporal and Complementary Learning

In this article, we study the problem of video-based action recognition. We improve the action recognition performance by finding an effective temporal and appearance representation. For capturing the temporal representation, we introduce two temporal learning techniques for improving long-term temporal information modeling, specifically Temporal Relational Network and Temporal Second-Order Pooling-based Network. Moreover, we harness the representation using complementary learning techniques, specifically Global-Local Network and Fuse-Inception Network. Performance evaluation on three datasets (UCF101, HMDB-51, and Mini-Kinetics-200) demonstrated the superiority of the proposed framework compared to the 2D Deep ConvNets-based state-of-the-art techniques.

Time in Associative Learning: A Review on Temporal Maps

Ability to recall the timing of events is a crucial aspect of associative learning. Yet, traditional theories of associative learning have often overlooked the role of time in learning association and shaping the behavioral outcome. They address temporal learning as an independent and parallel process. Temporal Coding Hypothesis is an attempt to bringing together the associative and non-associative aspects of learning. This account proposes temporal maps, a representation that encodes several aspects of a learned association, but attach considerable importance to the temporal aspect. A temporal map helps an agent to make inferences about missing information by applying an integration mechanism over a common element present in independently acquired temporal maps. We review the empirical evidence demonstrating the construct of temporal maps and discuss the importance of this concept in clinical and behavioral interventions.