Papers › Exploiting Optical Flow Guidance for Transformer-Based Video Inpainting

Exploiting Optical Flow Guidance for Transformer-Based Video Inpainting

24 Jan 2023arXiv:2301.10048archive 2025-07-28

Kaidong Zhang, Jialun Peng, Jingjing Fu, Dong Liu

Transformers have been widely used for video processing owing to the multi-head self attention (MHSA) mechanism. However, the MHSA mechanism encounters an intrinsic difficulty for video inpainting, since the features associated with the corrupted regions are degraded and incur inaccurate self attention. This problem, termed query degradation, may be mitigated by first completing optical flows and then using the flows to guide the self attention, which was verified in our previous work - flow-guided transformer (FGT). We further exploit the flow guidance and propose FGT++ to pursue more effective and efficient video inpainting. First, we design a lightweight flow completion network by using local aggregation and edge loss. Second, to address the query degradation, we propose a flow guidance feature integration module, which uses the motion discrepancy to enhance the features, together with a flow-guided feature propagation module that warps the features according to the flows. Third, we decouple the transformer along the temporal and spatial dimensions, where flows are used to select the tokens through a temporally deformable MHSA mechanism, and global tokens are combined with the inner-window local tokens through a dual perspective MHSA mechanism. FGT++ is experimentally evaluated to be outperforming the existing video inpainting networks qualitatively and quantitatively.

PaperPDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

hitachinsk/fgt mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Flow-Guided Transformer for Video Inpainting (arXiv:2208.06768; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 2 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
hitachinsk/isvi mentioned on GitHubpytorchSyntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Optical Flow EstimationVideo Inpainting

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Inpainting DAVIS FGT++ LPIPS (object) 0.035 #10 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++ LPIPS (square) 0.028 #10 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++ PNSR (object) 35.61 #10 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++ SSIM (object) 0.961 #10 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++ SSIM (square) 0.971 #10 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++* LPIPS (object) 0.027 #11 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++* LPIPS (square) 0.022 #11 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++* PNSR (object) 35.9 #11 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++* SSIM (object) 96.8 #11 of 11 Archive leaderboard report
Video Inpainting DAVIS FGT++* SSIM (square) 97.6 #11 of 11 Archive leaderboard report
Video Inpainting YouTube-VOS FGT++ LPIPS 0.025 #1 of 2 Archive leaderboard report
Video Inpainting YouTube-VOS FGT++ PSNR 35.02 #1 of 2 Archive leaderboard report
Video Inpainting YouTube-VOS FGT++ PSNR (square) 33.18 #1 of 2 Archive leaderboard report
Video Inpainting YouTube-VOS FGT++ SSIM 97.6 #1 of 2 Archive leaderboard report
Video Inpainting YouTube-VOS FGT++* LPIPS 0.022 #2 of 2 Archive leaderboard report
Video Inpainting YouTube-VOS FGT++* PSNR 35.36 #2 of 2 Archive leaderboard report
Video Inpainting YouTube-VOS FGT++* PSNR (square) 33.72 #2 of 2 Archive leaderboard report
Video Inpainting YouTube-VOS FGT++* SSIM 97.8 #2 of 2 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Inpainting

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections