DeOverlay:影片疊加移除基準與來源錨定流模型
DeOverlay: A Benchmark and Source-Anchored Flow for Video Overlay Removal
DeOverlay 以來源錨定流匹配移除影片中的字幕、標題和標誌等疊加內容,並保留場景結構。它提出 DVTE-Bench,涵蓋 1,631 組配對影片及八類文字疊加;DeOverlay 在五項重建與感知指標上均勝過四個基線,Mask LPIPS 和 Mask VFID 較最佳受測基線分別降低 58.3% 和 70.3%。
Published on Oct 19
Authors:
,
,
,
,
,
Abstract
DeOverlay uses source-anchored flow matching and visibility-normalized temporal aggregation to remove video overlays while preserving scene structure, and introduces DVTE-Bench for standardized evaluation.
Video overlay removal aims to recover scenes obscured by subtitles, titles, and logos while preserving surrounding content. Existing methods use temporal propagation or source-conditioned generation, but can leave residual overlays and distort scene structure when useful source information is obscured. We propose DeOverlay, a source-anchored flow model that reconstructs the clean video from the source latent with fixed source guidance. Specifically, we train a shared velocity network along paired source-to-clean paths using region-weighted flow matching, enabling successive cor- rections from intermediate states. To select local source cues, we develop visibility-normalized temporal aggregation that excludes masked positions and averages the remaining observations. Global source context and edit-region embeddings complement these cues through additive condi- tioning, retaining a single video-token sequence. To evaluate scene recovery across diverse overlay conditions, we introduce DVTE-Bench, comprising 1,631 paired videos across eight text-overlay categories, with OCR-screened clean targets, frame-aligned pixel masks, and a common evaluation protocol. Experiments on the complete release show that DeOverlay outperforms four baselines on all five reconstruction and perceptual metrics, reducing Mask LPIPS and Mask VFID by 58.3% and 70.3%, respectively, relative to the strongest evaluated baseline. Additional qualitative examples demonstrate the removal of graphic logos and stickers. We will make the complete benchmark, evaluation code, and test outputs availabl. Project page:https://yahooo-m.github.io/DVTE-Bench/
View arXiv page View PDF Add to collection
Get this paper in your agent:
hf papers read 2610.26081
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.26081 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.26081 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.26081 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co