CVJul 7, 2025

TLB-VFI: Temporal-Aware Latent Brownian Bridge Diffusion for Video Frame Interpolation

arXiv:2507.04984v15 citationsh-index: 4
Originality Highly original
AI Analysis

This addresses the efficiency and temporal modeling limitations in video frame interpolation for computer vision applications, representing a strong incremental improvement over existing diffusion-based methods.

The paper tackles video frame interpolation by proposing TLB-VFI, an efficient video-based diffusion model that extracts temporal information through 3D-wavelet gating and a temporal-aware autoencoder, achieving a 20% improvement in FID on challenging datasets and reducing parameters by 3x with 2.3x speedup.

Video Frame Interpolation (VFI) aims to predict the intermediate frame $I_n$ (we use n to denote time in videos to avoid notation overload with the timestep $t$ in diffusion models) based on two consecutive neighboring frames $I_0$ and $I_1$. Recent approaches apply diffusion models (both image-based and video-based) in this task and achieve strong performance. However, image-based diffusion models are unable to extract temporal information and are relatively inefficient compared to non-diffusion methods. Video-based diffusion models can extract temporal information, but they are too large in terms of training scale, model size, and inference time. To mitigate the above issues, we propose Temporal-Aware Latent Brownian Bridge Diffusion for Video Frame Interpolation (TLB-VFI), an efficient video-based diffusion model. By extracting rich temporal information from video inputs through our proposed 3D-wavelet gating and temporal-aware autoencoder, our method achieves 20% improvement in FID on the most challenging datasets over recent SOTA of image-based diffusion models. Meanwhile, due to the existence of rich temporal information, our method achieves strong performance while having 3times fewer parameters. Such a parameter reduction results in 2.3x speed up. By incorporating optical flow guidance, our method requires 9000x less training data and achieves over 20x fewer parameters than video-based diffusion models. Codes and results are available at our project page: https://zonglinl.github.io/tlbvfi_page.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes