Space Time Recurrent Memory Network

Understand

Transformers have recently been popular for learning and inference in the spatial-temporal domain.

• However, their performance relies on storing and applying attention to the feature tensor of each frame in video.
• Hence, their space and time complexity increase linearly as the length of video grows, which could be very costly for long videos.
• We propose a novel visual memory network architecture for the learning and inference problem in the spatial-temporal domain.