Learning Representations from Audio-Visual Spatial Alignment
This addresses the limitation of prior methods that ignore spatial cues in audio-visual representation learning, offering improvements for tasks such as action recognition and segmentation.
The paper tackles the problem of learning representations from audio-visual content by introducing a self-supervised pretext task for audio-visual spatial alignment using 360° video and spatial audio, and demonstrates its advantages on downstream tasks like action recognition and video semantic segmentation.
We introduce a novel self-supervised pretext task for learning representations from audio-visual content. Prior work on audio-visual representation learning leverages correspondences at the video level. Approaches based on audio-visual correspondence (AVC) predict whether audio and video clips originate from the same or different video instances. Audio-visual temporal synchronization (AVTS) further discriminates negative pairs originated from the same video instance but at different moments in time. While these approaches learn high-quality representations for downstream tasks such as action recognition, their training objectives disregard spatial cues naturally occurring in audio and visual signals. To learn from these spatial cues, we tasked a network to perform contrastive audio-visual spatial alignment of 360° video and spatial audio. The ability to perform spatial alignment is enhanced by reasoning over the full spatial content of the 360° video using a transformer architecture to combine representations from multiple viewpoints. The advantages of the proposed pretext task are demonstrated on a variety of audio and visual downstream tasks, including audio-visual correspondence, spatial alignment, action recognition, and video semantic segmentation.