YOSE: You Only Select Essential Tokens for Efficient DiT-based Video Object Removal

TL;DR

YOSE is an efficient fine-tuning framework for DiT-based video object removal that selectively processes essential tokens, significantly reducing inference latency while maintaining high visual quality.

Contribution

YOSE introduces Batch Variable-length Indexing and DiffSim modules to adaptively select tokens and simulate diffusion, enabling mask-aware acceleration in video object removal.

Findings

YOSE achieves up to 2.5X speedup in 70% of cases.

Inference time scales linearly with masked regions.

Maintains visual quality comparable to baseline methods.

Abstract

Recent advances in Diffusion Transformer (DiT)-based video generation technologies have shown impressive results for video object removal. However, these methods still suffer from substantial inference latency. For instance, although MiniMax Remover achieves state-of-the-art visual quality, it operates at only around 10FPS, primarily due to dense computations over the entire spatiotemporal token space, even when only a small masked region actually requires processing. In this paper, we present YOSE, You Only Select Essential Tokens, an efficient fine-tuning framework. YOSE introduces two key components: Batch Variable-length Indexing (BVI) and Diffusion Process Simulator (DiffSim) Module. BVI is a differentiable dynamic indexing operator that adaptively selects essential tokens based on mask information, enabling variable-length token processing across samples. DiffSim provides a diffusion process approximation mechanism for unmasked tokens, which simulates the influence of unmasked regions within DiT self-attention to maintain semantic consistency for masked tokens. With these designs, YOSE achieves mask-aware acceleration, where the inference time scales approximately linearly with the masked regions, in contrast to full-token diffusion methods whose computation remains constant regardless of the mask size. Extensive experiments demonstrate that YOSE achieves up to 2.5X speedup in 70% of cases while maintaining visual quality comparable to the baseline. Code is available at: https://github.com/Wucy0519/YOSE-CVPR26.