Strumming to the Beat: Audio-Conditioned Contrastive Video Textures
This work addresses video synthesis challenges for applications in entertainment and media by improving on classic methods with modern self-supervised techniques, though it is incremental in combining existing contrastive learning with video textures.
The paper tackles the problem of generating infinite video textures from a single video by learning frame representations and transition probabilities via contrastive learning, enabling audio-conditioned synthesis without fine-tuning and achieving better human perceptual scores than baselines.
We introduce a non-parametric approach for infinite video texture synthesis using a representation learned via contrastive learning. We take inspiration from Video Textures, which showed that plausible new videos could be generated from a single one by stitching its frames together in a novel yet consistent order. This classic work, however, was constrained by its use of hand-designed distance metrics, limiting its use to simple, repetitive videos. We draw on recent techniques from self-supervised learning to learn this distance metric, allowing us to compare frames in a manner that scales to more challenging dynamics, and to condition on other data, such as audio. We learn representations for video frames and frame-to-frame transition probabilities by fitting a video-specific model trained using contrastive learning. To synthesize a texture, we randomly sample frames with high transition probabilities to generate diverse temporally smooth videos with novel sequences and transitions. The model naturally extends to an audio-conditioned setting without requiring any finetuning. Our model outperforms baselines on human perceptual scores, can handle a diverse range of input videos, and can combine semantic and audio-visual cues in order to synthesize videos that synchronize well with an audio signal.