Fusion of Short-term and Long-term Attention for Video Mirror Detection

TL;DR

This paper introduces a novel video mirror detection method that combines short-term and long-term attention mechanisms to leverage both appearance and temporal context, achieving state-of-the-art results on a new challenging dataset.

Contribution

The paper proposes a fusion approach of short-term and long-term attention modules for improved video mirror detection, addressing the limitations of single-frame methods.

Findings

Achieves state-of-the-art performance on the new benchmark dataset.

Effectively leverages temporal information for more accurate mirror detection.

Demonstrates robustness across diverse video scenarios.

Abstract

Techniques for detecting mirrors from static images have witnessed rapid growth in recent years. However, these methods detect mirrors from single input images. Detecting mirrors from video requires further consideration of temporal consistency between frames. We observe that humans can recognize mirror candidates, from just one or two frames, based on their appearance (e.g. shape, color). However, to ensure that the candidate is indeed a mirror (not a picture or a window), we often need to observe more frames for a global view. This observation motivates us to detect mirrors by fusing appearance features extracted from a short-term attention module and context information extracted from a long-term attention module. To evaluate the performance, we build a challenging benchmark dataset of 19,255 frames from 281 videos. Experimental results demonstrate that our method achieves state-of-the-art performance on the benchmark dataset.