OcuFlow
An eye-adaptive pixel-level MeanFlow image inpainting framework for eyeglass reflection removal. It combines geometry-adaptive representation with a one-step pixel-level Mean Flow (pMF) mechanism, supporting native-resolution frequency-preserving synthesis for this specialized restoration task.
Timeline (last 90 days)
A study published on arXiv proposes OcuBench, a multi-source benchmark dataset, and OcuFlow, an eye-adaptive pixel-level MeanFlow framework, for addressing the problem of eyeglass reflection removal.
OcuFlow combines geometry-adaptive representation with one-step pixel-level MeanFlow (pMF).
OcuFlow employs native-resolution frequency-preserving synthesis to retain reliable observed details in the image while removing reflections.
In a blind user study, OcuFlow achieved a selection rate of 67.32%, which is 6.2 times the share of the second-best method.
Experiments show that OcuFlow demonstrates consistent advantages across three dimensions: reflection removal quality, ocular fidelity, and efficiency.
The research team stated that the code and datasets for both OcuBench and OcuFlow will be publicly released.
All Facts (6)
OcuFlow combines geometry-adaptive representation with one-step pixel-level MeanFlow (pMF).
50%UnverifiedOcuFlow employs native-resolution frequency-preserving synthesis to retain reliable observed details in the image while removing reflections.
50%UnverifiedIn a blind user study, OcuFlow achieved a selection rate of 67.32%, which is 6.2 times the share of the second-best method.
50%UnverifiedExperiments show that OcuFlow demonstrates consistent advantages across three dimensions: reflection removal quality, ocular fidelity, and efficiency.
50%UnverifiedThe research team stated that the code and datasets for both OcuBench and OcuFlow will be publicly released.
50%UnverifiedA study published on arXiv proposes OcuBench, a multi-source benchmark dataset, and OcuFlow, an eye-adaptive pixel-level MeanFlow framework, for addressing the problem of eyeglass reflection removal.
50%