Frame Interpolation with Consecutive Brownian Bridge Diffusion

TL;DR

This paper introduces a novel Brownian Bridge diffusion method for video frame interpolation that reduces variance in latent representations, enabling more deterministic and accurate intermediate frame synthesis, achieving state-of-the-art results.

Contribution

The paper proposes a unique consecutive Brownian Bridge diffusion approach that minimizes latent variance, improving deterministic frame interpolation in diffusion-based VFI models.

Findings

Reduces variance in latent representations for deterministic outputs.

Achieves state-of-the-art performance in video frame interpolation.

Enhances the stability and accuracy of diffusion-based VFI methods.

Abstract

Recent work in Video Frame Interpolation (VFI) tries to formulate VFI as a diffusion-based conditional image generation problem, synthesizing the intermediate frame given a random noise and neighboring frames. Due to the relatively high resolution of videos, Latent Diffusion Models (LDMs) are employed as the conditional generation model, where the autoencoder compresses images into latent representations for diffusion and then reconstructs images from these latent representations. Such a formulation poses a crucial challenge: VFI expects that the output is deterministically equal to the ground truth intermediate frame, but LDMs randomly generate a diverse set of different images when the model runs multiple times. The reason for the diverse generation is that the cumulative variance (variance accumulated at each step of generation) of generated latent representations in LDMs is large. This makes the sampling trajectory random, resulting in diverse rather than deterministic generations. To address this problem, we propose our unique solution: Frame Interpolation with Consecutive Brownian Bridge Diffusion. Specifically, we propose consecutive Brownian Bridge diffusion that takes a deterministic initial value as input, resulting in a much smaller cumulative variance of generated latent representations. Our experiments suggest that our method can improve together with the improvement of the autoencoder and achieve state-of-the-art performance in VFI, leaving strong potential for further enhancement.